XYO and Theta Just Solved AI’s Accountability Problem
AI & Infrastructure
XYO and Theta Just Solved AI’s Accountability Problem
For the first time, AI agents running on decentralized infrastructure can generate independent, on-chain proof of their own performance — thanks to a new integration between XYO Layer One and Theta EdgeCloud. No centralized cloud has ever offered this.
Crypto Coin Show·News & Analysis·May 2026
Autonomous AI agents are already operating at scale — initiating transactions, consuming cloud services, and coordinating with other agents, all without a human in the loop. But until now, there has been no reliable, independent way to verify whether those agents actually performed as intended. When no human is watching, the record of what happened — if it exists at all — lives inside the same system that ran the workload.
That changes today. XYO Layer One and Theta EdgeCloud have announced a new integration that gives AI agents something they have never had: verifiable, tamper-proof proof-of-performance written to a public blockchain by an independent observer.
Why This Matters Now
AI agents are already running production workloads for the Houston Rockets, Olympique de Marseille, and partners across the MLS, NBA, NHL, and Ligue 1 on Theta EdgeCloud. As agentic AI scales across industries, the question of accountability — did it actually do what it said? — becomes critical infrastructure, not a nice-to-have.
XYO watches EdgeCloud from the outside — independently measuring real uptime, speed, and reliability — and writes permanent records to XYO Layer One. Those records are stored in XYO Data Lakes and can be queried by anyone, at any time. The data never touches the system it is measuring, which is precisely the point.
Centralized clouds can’t offer this — their performance data never leaves their own systems.
What AI Agents Can Now Do That They Couldn’t Before
This integration unlocks a new capability for the broader AI industry: agents can now generate cryptographic receipts of their own execution. Every task completed, every service consumed, every coordination event — recorded independently and permanently on-chain, paid for in $XL1 with a portion of each payment burned.
For developers and enterprises deploying AI agents, this means auditability without trusting the platform running the agents. For end users interacting with AI systems, it means a publicly accessible performance history. For the broader Web3 ecosystem, it means the accountability layer that autonomous AI has been missing is now real.
Agents can now prove execution
Every task, coordination event, and service call can generate an on-chain record — written by an independent observer, not the agent itself.
Infrastructure can be audited publicly
Uptime, speed, and reliability records for Theta EdgeCloud are now permanently stored in XYO Data Lakes, open to anyone.
Builders can ship in days, not months
The XYO AI SDK compresses what would traditionally take months of engineering into a days-long integration, deployed directly to XYO Layer One.
The standard is open
Theta is the first partner — but XYO and Theta Labs are building a shared standard any infrastructure provider can adopt.
A New Accountability Layer for Agentic AI
The broader significance of this integration goes beyond XYO and Theta. As AI agents take on more consequential tasks — managing funds, executing trades, coordinating supply chains — the question of whether they performed correctly becomes a legal and commercial necessity, not just a technical curiosity.
Today’s announcement establishes a blueprint: an independent, blockchain-based verification layer that operates outside the systems it monitors. Theta EdgeCloud is the first infrastructure partner to ship a verification standard built this way. The XYO team says it will not be the last.
Token Mechanics: Every Verification Is a Transaction
For $XL1 and $XYO holders, the integration ties real-world AI activity directly to on-chain economics. Every record written to XYO Layer One is a transaction. Every transaction burns a portion of $XL1. As AI agent workloads scale across EdgeCloud and beyond, that activity flows through XYO Layer One — creating deflationary pressure anchored to actual infrastructure usage, not speculation.
$XL1 Token Flow
Agent Runs
AI workload executes on Theta EdgeCloud
→
XYO Verifies
Independent record written to XYO L1
→
$XL1 Paid
Every record is a transaction
→
Burn
Portion of every payment permanently removed
What Comes Next
The XYO AI SDK is now open for early access testing. Developers can begin building verified AI applications and agents — with deployment directly to XYO Layer One — starting today. Theta EdgeCloud is live as the first verified partner, with more integrations expected to follow as the standard gains traction.
Two of crypto’s earliest protocols. One accountability standard built for the age of autonomous AI.
Early Access — Now Open
Build with the XYO AI SDK
Start shipping verified AI apps and agents in days — not months.
Bittensor price predictions anticipate a high of $473.94 by the end of 2026.
In 2028, TAO will range between $842.56 and $1,000.54, with an average price of $921.55.
In 2032, TAO will range between $1,895.75 and $2,053.73, with an average price of $1,974.74.
Bittensor is one of the most renowned AI-facilitated decentralized networks that promotes blockchain and artificial intelligence infusion. By leveraging Proof of Learning (POL) technology, Bittensor supports user privacy while minimizing errors. The AI models within the network are reliable, flexible, and up-to-date with modern technological advancements. The AI-based Bittensor network prioritizes cross-chain integration and native token expansions to promote collaboration among various decentralized AI networks.
TAO uses reliable authentication methods to ensure a successful transfer of nodes through its AI knowledge to correct models. The process is made possible through the PoL consensus method, which secures this process. Moreover, this technology helps to develop different stages of more advanced AI technology within the blockchain. Bittensor also uses its TAO token to incentivize node operators and AI developers.
What’s next for Bittensor and TAO in 2026 and beyond? Let’s get into the TAO price prediction and technical analysis.
Overview
Cryptocurrency
Bittensor
Ticker
TAO
Current price
$277.96 (+0.71%)
Market cap
$3.04B
Trading volume (24-hour)
$171.08M
Circulating supply
10.93M TAO
All-time low
$30.40 on May 14, 2023
All-time high
$767.68 on Apr 11, 2024
24-hour low
$280.10
24-hour high
$269.05
TAO price prediction: Technical analysis
Metric
Value
Price Volatility (30-day variation)
8.04% (Very High)
14-day RSI
48.47
50-day SMA
$275.59
200-day SMA
$258.86
Market Sentiment
Neutral
Fear and greed index
30 (Fear)
Green days
13/30 (43%)
Bittensor price analysis
TL;DR Breakdown:
TAO price analysis confirms a bullish trend at $277.96.
The altcoin has gained 0.71% over the day.
TAO token has support at $265.
On May 25, 2026, TAO price analysis indicates a mild bullish daily trend, with Bittensor currently trading at $277.96. The altcoin has shown a 0.71% increase in value over the last 24 hours, primarily due to the swift recovery observed over the day, as the token is slipping from sellers’ grip. Buyers remain in control as the TAO price has the nearest support at the $274 level, and it may continue to maintain above the aforementioned level for the coming days.
TAO/USD 1-day chart analysis
The one-day price chart of Bittensor confirmed a bullish trend for the altcoin. The TAO/USD pair value has slightly recovered to $277.96 following a bearish spell. The comparatively high volatility suggests a higher chance of a reversal in the trend or further price appreciation.
The distance between the Bollinger Bands determines the market volatility. Currently, this distance is wide, leading to high volatility levels. Moreover, the upper limit of the Bollinger Bands indicator, indicating resistance, has shifted to $329. Whereby its lower limit, indicating support, has moved to a low of $247.
The Relative Strength Index (RSI) indicator is in the neutral region, in contrast to the other technical factors, which also seem to be bearish. Its curve also increased to 54 during the day. This slowly increasing price movement today reflects a relatively balanced trading setup in the market under the larger bearish trend. However, if the bullish momentum accelerates, the RSI value will move further up into the neutral region.
TAO/USD 4-hour chart analysis
The four-hour price chart for the Bittensor coin signifies a weak bearish trend, as the token’s price movements are again in a downward direction, with sellers trying to control the market. In the past few hours, the cryptocurrency’s value has slightly decreased to $277.81. Red candlesticks on the price chart signal a returning selling pressure.
The Bollinger Bands are expanding, as the volatility level is high on the 4-hour chart. The high volatility suggests higher market unpredictability. The upper Bollinger Band has shifted to a $289 high, indicating the resistance level. Conversely, the lower Bollinger Band is at $259, indicating the support level.
Multiple technical quantitative indicators are still bearish, but the RSI indicator is in the neutral region. However, the current score of 53 and decreasing numbers confirm selling pressure. The declining curve on the indicator’s graph shows rising selling activity and bearish progress as the market conditions turn unfavorable on an hourly basis.
Bittensor technical indicators: Levels and actions
Daily simple moving average (SMA)
Period
Value ($)
Action
SMA 3
273.13
BUY
SMA 5
274.28
BUY
SMA 10
271.05
BUY
SMA 21
289.05
SELL
SMA 50
275.59
BUY
SMA 100
255.45
BUY
SMA 200
258.86
BUY
Daily exponential moving average (EMA)
Period
Value ($)
Action
EMA 3
274.73
BUY
EMA 5
273.97
BUY
EMA 10
276.07
BUY
EMA 21
279.09
SELL
EMA 50
275.26
BUY
EMA 100
267.74
BUY
EMA 200
274.04
BUY
What can we expect from Bittensor price analysis next?
Bittensor (TAO) fundamental analysis indicates a bullish outlook for current market trends. The TAO/USD price has slightly recovered to $277.96, but the bearish momentum has not faded yet. Most technical indicators signal bearishness, but the price charts lean in favor of the buyers, suggesting a potential move toward the $294 level, due to today’s recovery.
Is Bittensor TAO a good investment?
TAO coin continues to trade higher, indicating growing adoption among crypto investors as AI development and machine learning progress. Despite this, the coin faces uncertainties and volatility like all other cryptocurrencies. Our Cryptopolitan price prediction explores its potential profit and expected movements from 2026 to 2032 while considering the past performance. However, this is not investment advice, and one must conduct their own research before taking any investment decision according to their risk tolerance.
Why is TAO up?
TAO is up primarily due to buying pressure from traders after some degree of bullish price action, mainly due to strong market sentiment surrounding speculative AI tokens and the AI industry at large. Recent stability near key support levels also played a role in the resurrection of the bullish trend as traders started buying following a bull rally, and the token’s price has also increased during the past 24 hours.
How much is the Bittensor stock worth?
Bittensor (TAO) powers the Bittensor Network and is not a stock. Stocks are usually traded on stock exchanges, and stock ownership represents a stake in a company. Buying TAO tokens gives the buyer certain rights within the Bittensor Network, for example, governance participation but not ownership in a company. However, Bittensor (TAO) tokens can be purchased and traded on different exchanges, including Binance, Bitget, Coinbase, KuCoin, and Kraken. See our price analysis part for day-to-day price changes of the TAO token.
What is the price prediction for TAO 2026?
The highest Bittensor (TAO) price prediction for 2026 is around $570.20, but it is not easy to predict Bittensor price movements due to its volatile nature.
Will Bittensor reach $1000?
Yes, Bittensor should surpass $1000 by 2028. Its price will range between $842.56 and $1,000.54 during that period, which makes it a viable option to buy Bittensor tokens, considering the future performance and long-term trends, as decentralized AI development is expected to scale exponentially.
What is the total supply of Bittensor?
The total supply of Bittensor (TAO) tokens is 21 million TAO.
Does Bittensor have a good long-term future?
According to most market observers, Bittensor TAO will trade higher in the coming years. However, factors like market crashes or difficult regulations could invalidate this bullish theory.
Recent news/ opinions on Bittensor
ORO just went live on Bittensor, a specialized marketplace focused on AI agents for real-world e-commerce. It is important to remember that ORO (subnet 15) outperformed OpenAI’s GPT-5.4 on some complex online shopping evaluations.
ORO is now live on Bittensor, an open arena for AI agents.
A decentralized evaluation platform where miners submit AI shopping agents and validators assess them independently, with top performers earning emissions.
— xTAO – a Bittensor company (@xtaohq) May 1, 2026
Bittensor price prediction May 2026
A break of resistance will result in a mini bull run, with the next target at $352 during the month. The average price is expected to be $263, according to the current forecast. In a bearish scenario, TAO could drop to $197 at its lowest.
Month
Potential low
Potential average
Potential high
May 2026
$197
$263
$352
Bittensor price prediction 2026
The technical indicators are bullish on TAO for the end of 2026. It is anticipated to trade between $134 and $473.94, with an average price of $394.95, according to the Bittensor price prediction.
Year
Potential low
Potential average
Potential high
2026
$134
$394.95
$473.94
Bittensor price predictions 2027-2032
Year
Minimum Price
Average Price
Maximum Price
2027
$579.26
$658.25
$737.24
2028
$842.56
$921.55
$1,000.54
2029
$1,105.86
$1,184.84
$1,263.83
2030
$1,369.15
$1,448.14
$1,527.13
2031
$1,632.45
$1,711.44
$1,790.43
2032
$1,895.75
$1,974.74
$2,053.73
Bittensor’s price forecast 2027
TAO is expected to gain bullish momentum in 2027. According to the updated Bittensor forecast, the token will range between $579.26 and $737.24, with an average price of $658.25.
Bittensor price prediction 2028
The Bittensor outlook strengthens further in 2028. Analysts expect TAO to trade between $842.56 and $1,000.54, with an average yearly price of $921.55.
Bittensor TAO price prediction 2029
The 2029 Bittensor price prediction suggests TAO will move between a minimum of $1,105.86 and a maximum of $1,263.83, settling at an average price of $1,184.84 for the year.
Bittensor price prediction 2030
For 2030, Bittensor price predictions indicate a trading range from $1,369.15 to $1,527.13, with an average expected price of $1,448.14.
Bittensor crypto price prediction 2031
In 2031, Bittensor price prediction, TAO is projected to range between $1,632.45 and $1,790.43, with an average price of $1,711.44, which is quite higher than its current value.
Bittensor price prediction 2032
The Bittensor price prediction for 2032 places TAO between $1,895.75 and $2,053.73, with an average price of $1,974.74.
TAO market price prediction: Analysts’ TAO price forecast
Platform
2026
2027
Digitalcoinprice
$202.28
$224.87
Coincodex
$245.77
$245.77
Cryptopolitan’s Bittensor (TAO) price prediction
According to our predictions, TAO could recover to $473.94 by the end of December 2026. We expect TAO to maintain a trading range of $579.26-$737.24, with an average of $658.25 in 2027. Note that the predictions are not investment advice. Seek independent professional consultation or do your research.
TAO launched on March 6, 2023, at $93.4, but fell below its opening price within a week, sliding into the $76 range.
By early April, it had lost half its value, dropping to $47, and continued downward to its $30.83 low in May before slowly recovering to $63 by the end of the month.
The token climbed to $86.18 in July, just under its launch price, then pulled back again and traded near $54 through October.
Momentum returned in November, pushing TAO into the $95 range, showing continuous improvement, and then sharply to a peak of $379 on December 15, 2023.
TAO trended downward into early 2024 but surged to its all-time high of $757.60 in March. It quickly corrected to $522 in April and continued weakening through mid-year, reaching $216 in July.
A brief rebound to $357 faded again as the token slipped back toward the mid-$200s by late summer, as per the crypto market price history records.
Momentum returned in October, pushing TAO into the $660 range before cooling to $468, according to the historical price data. It climbed once more to $679 in November but ultimately closed 2024 at $440.69, as the broader crypto market turned bearish again.
TAO opened in 2025 at $439.73, peaked at $565 in January, and its price decreased to the $324 level in February, taking down the token’s market capitalization as the technical indicators turned bearish due to some fundamental factors.
In March, TAO dipped to the $259 mark and descended further to $228 in April; however, in May, it recovered to $467 as the Bittensor market revived.
In October, TAO observed its year’s lowest prices extending toward $200.44.
TAO opened trading in November at $506, lost 46% of its value, and closed the month at $269.11, while at the start of December, the coin was trading between $256.29 and $298.90.
At the start of January 2026, TAO was trading near the $223 range, as the market shifted towards the bearish side.
In March, TAO traded below the psychological level of $200, but it surged past $300 in the month of April. At the start of May, TAO is trending above the $260 range, as current market sentiment turned decidedly bullish once again.
Big Tech OwnsYour Compute.Here’s Who’sTaking It Back.
While Big Tech races to build ever-larger data centers, 80% of existing GPU capacity sits idle. io.net is betting that the future of AI compute looks nothing like the past.
The numbers coming out of the hyperscalers are staggering. An estimated $650 billion is being spent on AI data center infrastructure in 2026 alone, with Amazon, Microsoft Azure, and Google Cloud racing to stake out compute real estate across the United States and beyond. Headlines about planned campuses have become routine. So have the headlines about delays.
Grid constraints, community opposition, soaring construction costs, and permitting backlogs have pushed back roughly half of planned US data center openings. The irony is sharp: the industry most loudly declaring a compute shortage is struggling to build its way out of one.
But there is a more uncomfortable truth underneath the construction race. The data centers that already exist are chronically underused. Industry estimates suggest that around 80% of global GPU capacity goes unutilized at any given time. Compute workloads are spiky by nature. A company trains a model, then the chips sit. Inference traffic surges and then falls quiet. The infrastructure built for peak demand idles through the troughs.
“Instead of having to build lots of data centers all over the world constantly, we should be juicing the data centers we have more effectively.”
Jack Collier, CMO, io.net
It is this inefficiency, not just the cost, that io.net was built to address. The company aggregates spare GPU capacity from secondary data centers, mining operations, and consumer-grade hardware, pooling it into a single marketplace that anyone can access. Three providers — AWS, Azure, and Google Cloud — control roughly 70% of global compute. The remaining 30% is fragmented across thousands of secondary operators and consumer hardware. io.net connects that fragmented supply into a single, accessible network.
The Business Case
Under $2 an Hour for an H200. That Is Not a Typo.
The flagship claim io.net makes is cost. H200 GPUs, among the most powerful chips available for AI workloads, are listed on the io.net platform today for under $2 per hour. The same hardware on AWS or Google Cloud runs $25 to $30 per hour. For a startup burning 40 to 60 percent of its operating budget on compute, that difference is not marginal. It is existential.
H200 on io.net
<$2
per hour
H200 on AWS
$25–30
per hour
Devices live
10K+
across 138 countries
Cluster setup
~2 min
no waitlist, no KYC
Token
$IO
staked by suppliers
Leonardo.ai, the AI imaging company recently acquired by Canva, is perhaps io.net’s most prominent case study. The team uses io.net for inference workloads and has credited the cost savings with giving them room to innovate faster. That kind of reference point matters when trying to convince web2 companies that decentralized infrastructure is not an experiment.
And that, according to io.net CMO Jack Collier, is where most of the company’s revenue actually comes from today. “Most of our revenue comes from web2,” he noted, “people who don’t even know that they’re building on crypto rails.” The blockchain layer, in other words, is infrastructure, not identity.
Why Web2 Companies Aren’t Switching Faster
Lock-in is real. Once a business has built its stack on AWS or Azure, the connective tissue runs deep through every service, billing integration, and workflow. Extraction is costly and disruptive. Add to that the narrative pressure from hyperscalers themselves, who have significant marketing budgets dedicated to reinforcing fears of GPU shortages, and the inertia becomes easier to understand. io.net’s answer is to let the price differential speak for itself and build the track record one customer at a time.
Resilience and Geography
When AWS Goes Down, Everything Goes Down. That Is the Problem.
Centralized infrastructure carries a centralized failure mode. When a major cloud provider experiences an outage, the cascade is immediate and broad. Thousands of services, often unrelated to one another, go dark simultaneously because they all share the same dependency.
Decentralized compute inverts this logic. io.net customers can distribute their workloads across GPU clusters in four or five countries simultaneously. If one node fails, traffic reroutes. For global products, this also enables something else: local inference. A company serving customers in Japan can run its models from Japan. Customers in South Africa get inference from South Africa. Latency drops. Performance improves. The infrastructure adapts to geography rather than forcing geography to adapt to infrastructure.
This geographic flexibility, available today across more than 138 countries, is one of io.net’s less-discussed advantages. It quietly solves a problem that hyperscalers solve only expensively and slowly, by building new regional data centers.
Full Interview — CCS Blockchain Interviews
Jack Collier, CMO of io.net, speaks with Ashton Addison of the Crypto Coin Show about decentralized compute, the IDE, Agent Cloud, and the future of AI infrastructure.
Fixing the Economics
The Incentive Dynamic Engine: From Inflation to Utility
Most decentralized physical infrastructure networks, DePIN projects in crypto parlance, share a structural problem. They incentivize suppliers by minting new tokens and distributing them as rewards. When token prices rise, suppliers flood in. When prices fall, they leave. The network’s supply is held hostage to speculation rather than anchored to real demand.
io.net has responded with what it calls the Incentive Dynamic Engine, or IDE, scheduled for full implementation in Q2 2026. The shift is fundamental: instead of paying suppliers a fixed amount of IO tokens each month, suppliers are now compensated in proportion to actual demand on the network. Payments are denominated in USDC-equivalent value of IO, meaning suppliers receive stable dollar-value compensation regardless of token price fluctuations.
Revenue above what is needed to pay suppliers flows into a reserve vault. That vault absorbs volatility. In price downturns it subsidizes supplier rewards. In stronger markets, excess emissions from that vault are burned. io.net has committed to burning at least 50 percent of those excess emissions permanently, meaning the total IO supply contracts over time as the network grows.
IDE Change
Detail
Status
Network model
Supply-driven → demand-driven
Q2 2026
Supplier payments
USDC-equivalent IO (stable dollar value)
Q2 2026
Emissions burn
50% minimum of vault excess
Ongoing
Network direction
Inflationary → deflationary over time
By design
“Tokens aren’t just there as an investment vehicle. They’re there to power a trustless network.”
Jack Collier, CMO, io.net
The result is a tokenomic model where the value of IO is tied directly to the utility of the network it powers, not to sentiment cycles. For anyone evaluating whether a blockchain project is serious, that kind of alignment is among the clearest signals available.
The Agent Economy
AI Agents That Buy Their Own Compute
One of io.net’s more forward-looking product moves is Agent Cloud, launched in March 2026. The premise is simple and slightly startling: AI agents, which already automate enormous swaths of software work, can now autonomously purchase the compute power they need to run. No human in the loop. No approval workflow.
Launched
Mar 25
2026
Protocol
MCP
library by io.net
Payment
Both
crypto or fiat
Guardrails
Yes
spend limits built in
Agent Cloud is built on a Model Context Protocol library created by io.net. An agent with access to a wallet can query the io.net marketplace, identify the GPU configuration it needs, and complete the purchase automatically. Guard rails prevent runaway spending, with limits on how many devices can be acquired and for how long.
The concept points toward something larger. If AI agents are going to be first-class economic participants, they need infrastructure that is programmatically accessible. Centralized cloud providers require account creation, billing agreements, and human oversight at the procurement layer. A permissionless marketplace, accessible via API and payable in crypto or fiat, removes those friction points entirely.
“Our CEO talks quite passionately about a world where AI agents are being spun up themselves and are able to purchase their own compute power and run entirely autonomously,” Collier said. It is a vision of compute as a commodity that intelligent systems consume on demand, the same way applications consume electricity or bandwidth.
Where This Goes
The Demand Curve Only Runs One Direction
The case for decentralized compute rests on a straightforward projection: AI demand will grow faster than centralized infrastructure can be built, and the inefficiency of today’s capacity utilization leaves enormous room for networks that can aggregate and reallocate idle supply. io.net is not alone in making this argument, but it is among the furthest along in proving it with revenue.
From zero to $25 million in annualized revenue, in roughly a year of serious commercial operation, against a global data center market measured in the hundreds of billions, there is a long road ahead. But the trajectory is real, the product is live, and the customers are increasingly the kind of companies who do not think of themselves as crypto users at all.
That quiet expansion — blockchain as invisible infrastructure rather than explicit identity — may be the most durable growth story in the space. Spin up a cluster at io.net in two minutes. No waitlist. No KYC labyrinth. Just compute, available to whoever needs it.
The Hallucination Problem — And the Data Layer That Solves It
Analysis
AI Keeps Lying. The Fix Isn’t a Better Model. It’s Better Data.
OpenAI formally admitted it in 2025: language models are structurally rewarded to guess rather than say “I don’t know.” A growing field of blockchain-based data infrastructure projects believes the real fix happens before the model is ever trained—at the data layer itself.
April 25, 2026 · 11 min read · DePIN / AI / Blockchain / Data Infrastructure
The Problem
The Test That Punishes Honesty
Picture a standardized exam where leaving a question blank gives you zero, but a wrong guess gives you a chance at a point. Rational test-takers guess. Always. Now imagine your AI is trained on exactly that exam, at scale, across billions of questions. You haven’t built a truthful system—you’ve built an optimized guesser.
This isn’t a metaphor. It is the exact structural critique published by OpenAI researchers in September 2025. Their paper, Why Language Models Hallucinate, argues that models hallucinate because standard training and evaluation procedures reward guessing over acknowledging uncertainty. Saying “I don’t know” scores zero. Confidently guessing wrong at least has a chance of scoring something—so over thousands of benchmark questions, guessing pays.
“Hallucinations are not a mysterious artifact of neural networks. They are a predictable outcome of how we train and evaluate language models.”
The paper’s core insight is structural, not incidental: accuracy-only leaderboards dominate how the entire field evaluates models. On those scoreboards, a model that guesses boldly—and occasionally gets lucky—outranks a model that abstains with honest uncertainty. The scoring hasn’t fundamentally changed, so neither has the behavior.
⚠ The Core Mechanism
For a question a model doesn’t know—say, a specific person’s birthday—guessing “September 10” gives a 1-in-365 chance of being right. Saying “I don’t know” guarantees zero points. Multiplied across millions of training examples, the statistical pressure to guess is enormous. This is not a prompt engineering problem. It is baked into how AI is scored.
Scaling Hasn’t Fixed It
OpenAI acknowledges that GPT-5 significantly reduces hallucinations—especially in reasoning tasks—but still confirms they occur. The Vectara FaithJudge Leaderboard in 2025 put grounded hallucination rates at roughly 15–16% for GPT-4o and Claude 3.7 Sonnet, with Gemini 2.5 Flash around 6%. Those are meaningful improvements. They are not solutions. Even a 6% hallucination rate, considered excellent by benchmark standards, translates into serious operational errors at scale—a corrupted field in a medical record, a fabricated legal citation, a wrong fact embedded in a financial report.
More parameters did not fix this. Larger context windows did not fix this. The problem isn’t the model’s size—it’s the incentive to guess, which is written into the scoring system that shapes training.
§
Data Quality
Garbage In, Hallucinations Out
Training incentives are one dimension of the hallucination problem. Data quality is another—and arguably the deeper one. When models are trained on vast swaths of the internet, they absorb noise, contradictions, outdated facts, fabrications, and synthetic text at scale. The model has no reliable way to distinguish between a peer-reviewed paper and a confidently written blog post that happens to be completely wrong.
OpenAI researchers describe this as the GIGO problem—Garbage In, Garbage Out. Post-training techniques like RLHF can reduce errors like common misconceptions and conspiracy theories, but they cannot fundamentally undo what was baked in during pretraining. If the data layer is polluted, the model is polluted. No amount of fine-tuning fully reverses that.
The bias problem is equally serious. AI trained on data that skews toward wealthy, English-speaking markets will perform well in those markets and fail quietly everywhere else. Markus Levin of XYO has pointed to a concrete example: when the COIN App was translated into Amharic—spoken by 57 million people in Ethiopia—ChatGPT’s translations were riddled with errors. Not because the model was broken, but because Ethiopia was not a priority data market. The training signal simply wasn’t there.
✕ Status Quo Data Pipeline
Web-scraped text with no origin verification
No audit trail for when or how data was generated
Cannot distinguish fact from plausible fiction at scale
Skewed toward high-resource languages and markets
Models incentivized to guess when knowledge gaps appear
✓ Verified Data Pipeline
Cryptographic Proof of Origin for every data point
Immutable on-chain audit trail anchored to real events
Data tied to real-world sensors and independent validators
Decentralized collection across underserved geographies
AI can verify claims rather than pattern-match guesses
✓ What Verified Data Has Already Proven
AI trained on high-quality, verified, structured data has produced breakthroughs in science and academia that seemed impossible a decade ago—protein folding, drug discovery, climate modeling. The models evolved, but so did the data. The two are inseparable. Better data is not a nice-to-have for AI. It is the single highest-leverage input in the stack.
§
The Skeptics
The Case Against Blockchain as a Data Fix
Not everyone is convinced that decentralized infrastructure is the right answer. The counterarguments deserve a fair hearing.
Retrieval-Augmented Generation (RAG) already helps. Many AI deployments now use RAG pipelines—anchoring model responses to retrieved documents rather than relying on baked-in training data. This reduces hallucinations significantly in enterprise contexts. Stanford’s 2025 legal RAG reliability work showed meaningful gains in accuracy for grounded tasks. The argument: you don’t need a new blockchain; you need better retrieval architecture.
Benchmark reform may be sufficient. OpenAI’s own paper proposes a targeted fix: change how accuracy-only scoreboards are weighted so that abstaining scores better than confident wrong answers. If the field adopts uncertainty-aware evaluation, the training incentive to guess diminishes—without needing any new data infrastructure at all.
Blockchain adds complexity without guaranteeing quality. Cryptographic proof of origin tells you where data came from, not whether what was recorded was true. A network of nodes can corroborate a false reading just as easily as a true one if the sensors or participants are compromised. Garbage In, Garbage Out applies to DePIN systems too.
⚠ A Fair Critique
RAG, better benchmarks, and RLHF-based fine-tuning are all genuine improvements and are already reducing hallucination rates in production systems. None of them, individually or in combination, has yet eliminated the problem—particularly for low-resource languages, niche domains, and real-time physical data. That gap is where verified data infrastructure makes its case.
The honest answer is that these approaches are complementary, not competing. Better benchmarks fix the training incentive. RAG grounds outputs in documents. Verified on-chain data grounds those documents in reality. Each layer addresses a different failure mode. The question isn’t which one wins—it’s which combination gets AI closest to reliable.
§
The Field
Who Else Is Building the Data Layer
XYO is not alone in recognizing that verified, decentralized data infrastructure is a missing piece of the AI stack. A cluster of projects has been converging on the same thesis from different angles.
Ocean Protocol
Ocean Protocol focuses on data marketplaces—enabling individuals and organizations to publish, share, and monetize datasets while maintaining control. Its model addresses a different angle of the same problem: not just verifying data provenance, but creating economic incentives for high-quality data contributors to participate in the first place. For AI training, a well-structured data marketplace with verified provenance is a meaningful step toward cleaner inputs.
Chainlink
Chainlink’s oracle network is arguably the most battle-tested decentralized data verification layer in production. Its core function is bridging off-chain real-world data—price feeds, weather data, sports scores, financial events—onto blockchains in a tamper-resistant way. While Chainlink’s primary use case has been DeFi smart contracts, the infrastructure directly applies to AI: verified, real-time external data feeds that a model can query with known provenance rather than guessing from training data.
Filecoin & The Storage Layer
Filecoin approaches the problem from persistence: decentralized, verifiable storage of data at scale. If AI training datasets can be stored and retrieved from a verifiable, censorship-resistant layer, it becomes harder to corrupt or quietly alter training inputs over time. Combined with provenance tracking, decentralized storage is a foundational piece of any serious verified-data architecture.
Ceramic Network
Ceramic is building a decentralized data streaming protocol—a layer for mutable, user-controlled data that remains verifiable across applications. Where most blockchain data is static, Ceramic enables dynamic, updateable data streams with identity and provenance attached. For AI applications that need fresh, real-world signals rather than stale training snapshots, this is an important architectural piece.
✓ A Converging Thesis
These projects approach data verification from different angles—marketplaces, oracles, storage, streaming—but they share a core conviction: that unverified, centrally curated data is a structural weakness in the AI stack, and that decentralized infrastructure can address it in ways that no single company controlling its own data pipeline can. The field is early, fragmented, and competitive. But the direction is coherent.
// Case Study: How XYO Verifies Data
From Raw Reality to Verified Truth
S
Sentinel
IoT devices & smartphones gather raw real-world data — location, proximity, environmental signals
B
Bridge
Relays data across the network with cryptographic bound-witness interactions between nodes
A
Archivist
Stores verified data with immutable Proof of Origin & Proof of Location anchored to XYO Layer One
D
Diviner
Answers queries with provably verified facts — enabling AI to act on truth rather than statistical guesses
Case Study — XYO
XYO and the Case for Physical Data Verification
Among the projects building verified data infrastructure, XYO occupies a specific and significant niche: real-world physical data. Founded in 2018 as the first DePIN project, its network now spans over 10 million nodes—smartphones, IoT sensors, and edge devices—across nearly every country on earth. Each node participates in a process called bound witnessing, where multiple independent nodes corroborate the same physical event cryptographically, making any single data point extremely difficult to falsify.
In September 2025, XYO launched XYO Layer One—the first blockchain designed from the ground up for data-heavy industries. After seven years of operations, the team concluded no existing chain could handle the volume, latency, and validation requirements of real-world physical data at scale. So they built their own, with a dual-token model: $XYO for governance and staking, $XL1 for gas and transactions.
@OfficialXYO on X
“XYO is going to end AI hallucinations. OpenAI has admitted that LLMs are structurally rewarded for guessing, and will always hallucinate. Saying ‘I don’t know’ scores zero on a benchmark, so they don’t say it.¹ AI trained on real, verified data has already done things for science and academia that seemed impossible a decade ago.² Better AI is possible, and it hinges entirely on the quality of data underneath it. XYO is building that data layer for everyone. Hallucination isn’t unsolvable. It just hasn’t been solved for everyone yet. That changes with XYO.”
¹ “Why Language Models Hallucinate,” OpenAI · ² “Good AI Starts Before the Model. It Starts With the Data,” XYO
Real Deployments, Real Stakes
What distinguishes XYO from many blockchain infrastructure plays is that it has been generating revenue and real-world traction for years before the AI data conversation caught up to it. The network generated $8.8 million in revenue in 2024, with 80% of users outside the crypto ecosystem entirely. In March 2026, XYO partnered with climate analytics firm Resiliocs to provide cryptographic verification for environmental and geospatial data used in climate risk modeling—an application where data accuracy is legally and financially material. In December 2025, Revolut, with a reported $75 billion valuation, listed $XYO—the first major fintech to add a DePIN token.
What XYO is specifically good at—physical location verification, real-world event attestation, Proof of Origin for sensor data—is exactly the category of data that current AI training pipelines handle worst. Language models can approximate text. They cannot approximate the physical world. That is the gap XYO is positioned to fill.
✓ Where XYO Fits the Broader Picture
XYO is not trying to solve every dimension of the hallucination problem. It is building the physical data verification layer that projects like Ocean Protocol (marketplaces), Chainlink (oracles), and Filecoin (storage) don’t specialize in. In a mature verified-data ecosystem, these layers are complementary. XYO’s edge is the depth and scale of its physical-world node network—10 million strong, built over seven years, before anyone was calling it AI infrastructure.
§
Conclusion
The Data Layer Is the Unlock
The AI industry has spent enormous resources making models larger, smarter, and better at reasoning. Those investments have paid off. But the hallucination problem has persisted because its root cause was never primarily about model architecture. It was about incentive structures and data quality—two things that no amount of additional compute directly fixes.
Benchmark reform, RAG pipelines, and fine-tuning are real improvements that are already helping in production systems. But they operate on top of a data foundation that remains largely unverified, unaudited, and biased toward the markets that happened to generate the most internet text. That foundation is what the DePIN and blockchain data layer is trying to fix.
It is early. The ecosystem is fragmented. Chainlink, Ocean Protocol, Filecoin, Ceramic, and XYO are each approaching one corner of a large problem, and none of them has yet become the dominant infrastructure standard for AI data verification. That race is still open.
Reliable data is now the most valuable resource, yet the infrastructure to verify and process it on-chain still did not exist. That is why we built XYO Layer One from the ground up.
What is no longer early is the problem itself. Hallucination rates of 6–16% across frontier models are not acceptable for high-stakes applications. The costs—fabricated legal citations, corrupted medical data, biased outputs in underserved communities—are real and documented. The question for AI’s next decade is not whether models will get more capable. They will. The question is whether the data underneath them can be trusted.
The answer starts before the model. It starts with the data. And the infrastructure to verify that data at global scale is being built right now—by a field that, until recently, was mostly ignored by the AI conversation. That’s changing fast.
GrandCroix and Ambient Network Announce Collaboration to Launch First Native DEX for DePIN Ecosystem in Q2 2026
Press Release
GrandCroix and Ambient Network Announce Collaboration to Launch First Native DEX for DePIN Ecosystem in Q2 2026
The partnership will introduce a purpose-built decentralized exchange, native cross-chain bridge, and liquidity infrastructure for Ambient ecosystem
Miami, FL—April 2026
GrandCroix, a Miami-based new generation of mining company for decentralized AI networks, today announced a strategic collaboration with Ambient Network, to launch the first native decentralized exchange (DEX) for the Ambient ecosystem in Q2 2026.
The platform will ship with a full DeFi stack from day one – including an automated market maker (AMM), cross-chain bridging, and built-in liquidity tools – designed specifically for Ambient.xyz
Ambient is an SVM-compatible PoW L1 that will serve as a cornerstone of the agentic economy, unleashing Asimov-ian intelligence on chain. It is 10x more efficient than incumbent crypto AI systems, and features:
Fully verified Inferencewith <.1% overhead but high security on one huge smart model (600b+ parameters) and its fine-tunes
10x better training performancethan existing approaches
Extremely high utilization of minersdue to optimization on a single model for inference and validation
A non-blocking proof of work consensusthat foregrounds economic competition around the core activities of the network (inference, fine tuning, training) while maintaining extraordinary TPS.
Participants earning Ambient token rewards have no native trading venue, limited cross-chain access, and no way to provide liquidity or earn yield within the ecosystem they’re helping build.
To close all three gaps, GrandCroix is building a vertically integrated DeFi platform – not just a swap interface – engineered for the specific needs of DePIN participants. Built on Ambient and scheduled to go live in Q2 2026, the platform will launch with:
AMM Swap EngineOptimized for $AMB and core trading pairs (SOL, USDC, USDT), with support for concentrated liquidity positions.
Native Cross-Chain BridgeIntegrated bridge enabling $AMB transfers between Solana and EVM-compatible chains (Ethereum, Base, Arbitrum) from day one, with no third-party bridge required.
Liquidity Positioning DashboardTools for liquidity providers to set price ranges, monitor performance, and manage positions. Designed to be accessible to sensor operators, not just DeFi power users.
Rather than bolting together fragmented third-party tools, the platform delivers a unified on-chain experience tailored to the people actually participating in the network.
The timing reflects a structural shift: as DePIN ecosystems mature beyond hardware deployment into data monetization and token utility, they need native financial infrastructure to sustain growth. Without it, value leaks to centralized intermediaries and cross-chain friction discourages participation. Ambient’s institutional backing from a16z crypto and Delphi Ventures – signals that the network is ready for this layer.
Every blockchain needs a financial layer to enable builders to build on top of it and support its growth. Our collaboration with Ambient is about building that missing layer – so participants can not just contribute data, but actively participate in the ecosystem’s economic growth by building apps.
Nour De VosFounder & CEO of GrandCroix
Liquidity Strategy
GrandCroix will seed initial liquidity from its own treasury and mining operations, providing day-one trading depth across core pairs. Ambient will also provide additional liquidity to the dex funded t will support early liquidity growth.
Roadmap
Q2 2026
AMM, Bridge, and LP Dashboard live on Solana mainnet
Q3 2026
Additional bridge chains, limit orders, and programmatic trading API
Q4 2026
Community governance launch with on-chain voting for fee parameters and new pair listings
For Media Inquiries, contact press@grandcroix.com
About GrandCroix
GrandCroix is a decentralized AI and DeFi infrastructure company headquartered in Miami, Florida, and a division of Group NDV. The company operates at the intersection of decentralized computing and decentralized finance, building essential infrastructure for emerging blockchain ecosystems. GrandCroix’s operations span active mining across major decentralized networks — including Bittensor, NousResearch, Gonka AI, Ambient Network, and Psy Protocol — large-scale GPU infrastructure management, and native DeFi product development.
About Ambient Network
Ambient sets out to address the deficiencies just described by building a fundamental pillar of the agentic economy: an AI secured blockchain ten times more efficient than incumbent systems with built-in privacy and censorship resistance that runs a single, huge, highly performant, auditable, and transparent model (and its fine tunes) at low latency by leveraging hyperscale on-chain distributed computing to deliver human-like capabilities to applications on-chain and cross-chain.
XYO’s Layer 1 Gets 2–5x Faster as AI Demand for Verified Data Heats Up | Crypto Coin Show
Network Update·XYO Network·April 2026
XYO’s Layer 1 Gets 2–5x Faster as AI Demand for Verified Data Heats Up
XL1’s latest performance upgrade cuts block validation time significantly, as XYO Labs doubles down on its role as a data provenance layer for AI systems and physical infrastructure networks.
AA
Ashton Addison, Editor in Chief
Crypto Coin Show · April 2026
XYO Network
2–5×
Speed increase
~1mo
Dev time vs. 6 months prior
2018
CCS first covered XYO
Q3 ’25
XL1 mainnet launch
XYO Network has shipped a significant performance upgrade to XL1, its Layer 1 blockchain, delivering a 2 to 5 times improvement in block processing speed less than a year after the network’s Q3 2025 mainnet debut. The upgrade arrives as demand for verified, on-chain data provenance is accelerating across the AI sector — a use case XYO has been building toward since long before decentralized physical infrastructure networks became a mainstream narrative.
The improvements were developed in roughly one month, a timeline that would have required approximately six months under traditional development cycles. The team credits AI-assisted development tools with enabling that compression, using them for continuous performance profiling, unit test generation, and regression detection across builds.
What the Upgrade Actually Changes
The core speed gain comes from faster account balance indexing during block validation. When a node validates a new block, one of the more computationally expensive operations is retrieving current account balances. Improving how that data is indexed reduces the time it takes to finalize each block — and that improvement compounds across the whole chain.
For stakers, more blocks per unit of time means more XL1 rewards generated from staking XYO. It also means more XL1 burned through gas, which accelerates the protocol’s path from net-inflationary toward deflation — the point at which token burn from usage exceeds new issuance from block rewards.
How XL1 Token Economics Work
Users stake XYO to participate in the network and earn XL1 as a block reward. XL1 pays gas on the chain, and that gas is burned. Faster blocks means more XL1 issued to stakers and more burned through usage — tightening supply dynamics on both ends simultaneously.
The team also built in systematic regression testing: automated profiling benchmarks that compare performance against previous builds after each release, catching slowdowns before they accumulate. A chain that improves 10% in one build and regresses 10% in the next has not actually improved. Preventing that is a different engineering problem than raw optimization, and one the team now has automated tooling to address.
Why AI Needs What XYO Builds
XYO’s core infrastructure has always been about data provenance — the ability to record not just what a piece of data says, but where it came from, when it was captured, and by whom, in a way that is immutable and independently verifiable. That problem has become considerably more urgent as AI systems ingest larger volumes of data from uncontrolled sources.
Most large language models learn from the open internet, and the internet does not come with a chain of custody. Data may be inaccurate, synthetic, outdated, or derived from sources that were never authorized to share it. As AI becomes embedded in financial decisions, medical systems, and autonomous vehicles, the inability to audit what a model learned from is shifting from an academic concern to a legal and commercial one.
“Once provenance goes into a model, it becomes blurred and kind of lost. At some point they’re going to have to start having audit trails.”
Ari Trout, Co-founder & CEO, XYO Labs
XYO’s architecture separates raw data storage from the metadata required to verify it. The full dataset — a high-resolution sensor reading, a video frame, a precise location coordinate — lives off-chain in private storage. What goes on-chain is a hash, a timestamp, and a reduced version of the data: enough to prove the original existed at a specific time without exposing its contents. If the full data ever needs validation, it can be revealed and checked against the on-chain record. The team calls this “reveal on demand” privacy.
The Witness System: Micro-Consensus for Real-World Data
One of the more distinctive parts of XYO’s architecture is how it handles data from physical sources — IoT sensors, weather stations, GPS devices, cameras. Rather than trusting a single source, the protocol uses a witness-based system: multiple independent sources observe the same data point without knowing about each other, and their readings are compared. If four out of five agree and one is an outlier, the outlier gets flagged. If consensus is insufficient, the data is resampled.
This is micro-consensus applied to data collection rather than transaction validation — particularly relevant for the kinds of physical data AI systems are increasingly asked to act on, where a single compromised source could silently corrupt a model’s understanding of reality.
Storage Architecture
XL1 is deliberately designed to keep most data off-chain. Storing a 20-gigabyte video file on a blockchain is not practical, and XYO does not try to do it. Instead, the chain stores the provenance record — hash, timestamp, metadata — while the underlying data lives elsewhere. The principle mirrors how Ethereum’s off-chain indexers like Etherscan work: the source of truth is on-chain, but the full queryable picture is assembled by infrastructure that indexes against it.
Partners wanting higher-frequency data commits — more timestamps per second for a given sensor stream — can now do so at the faster block rate without degrading the rest of the network.
Physical AI and the Road Ahead
XYO tracked location data from its founding, and autonomous vehicles represent a natural next frontier. A self-driving car generates continuous streams of camera, LIDAR, and sensor data that must be distilled in real time by an edge device before any of it can be acted on. The car cannot stream raw video to a remote server and wait for instructions — decisions happen at the edge, from compressed metadata. But the original sensor data and a provenance record of how the edge device processed it still has audit value for liability, safety review, and model improvement.
The same logic applies to home robots, industrial drones, and any physical AI system where something going wrong generates a question: what did it actually do, and what was it seeing when it did it? XYO’s infrastructure is designed to answer that question without requiring every frame to live on a public ledger.
Developer Access
Developers can connect to XL1 now via browser wallet injection, a JavaScript SDK, or direct RPC endpoint calls. SDKs for Go, Kotlin, and additional languages are in development, with AI-assisted porting accelerating the timeline considerably compared to traditional manual porting.
AI as a Development Tool, Not Just a Use Case
The efficiency gains in XL1’s latest release were themselves produced using AI. The team used AI coding assistants to accelerate development, generate test coverage, run profiling benchmarks, and catch regressions that would otherwise require manual review. The result is a team that can iterate on its blockchain in weeks rather than quarters — a meaningful advantage in an environment where protocol development speed increasingly determines competitive position.
It is an example of what XYO’s co-founder describes as tooling mattering more than raw compute: giving a system the right tools to operate efficiently, rather than simply throwing more resources at it. The same principle that makes XYO relevant to AI data infrastructure is the one that made its latest upgrade possible.
Watch: Ari Trout on XYO Layer 1, AI Data Verification & DePIN
Full conversation with XYO Labs co-founder and CEO Ari Trout covering the XL1 performance upgrade, data provenance architecture, AI tooling, witness-based consensus, and the role of physical infrastructure in on-chain AI.
XYO Layer One Gets Faster, Stronger, and Ready for What’s Next
Protocol Update · XYO Network · Layer One
XYO Layer One Gets Faster, Stronger, and Ready for What’s Next
A throughput jump of up to 5x, dual DataLake support, and tightened validator stability — the latest round of updates from Arie Trouw signals a chain moving from foundation to momentum.
AA
Ashton Addison
Founder & CEO · Crypto Coin Show · Since 2014
11 April 2026
XYO Layer One has been running quietly beneath the surface — and with the latest round of updates shipped directly by co-founder, CEO, and CTO Arie Trouw, that quiet work is starting to show. More throughput, more flexible data architecture, and a more stable validator layer: these are not marketing milestones. They are the kind of changes that make everything built on top of the chain more reliable.
Trouw’s hands-on involvement throughout the build has been consistent since the start. The updates now arriving aren’t the output of a roadmap drafted in a boardroom — they reflect someone who has been shaping the architecture directly. That context matters when reading what has changed.
5x
Throughput improvement (upper range)
2
DataLake modes — private & public
XL1
Float still low — window remains open
I.The Updates
Four Changes That Actually Matter
The headline number is throughput. Chain capacity has been pushed significantly higher — landing in the range of two to five times the previous ceiling. That range isn’t vagueness: it reflects how performance improvements compound differently depending on load. Under pressure, the gains are most visible. Data-heavy use cases that previously hit artificial ceilings now have room to run.
Alongside raw capacity, the team has added both private and public DataLake support directly into the SDK. The significance of this is architectural. Developers can now decide at the implementation level how data is stored, accessed, and shared — permissioned and controlled for sensitive records, open and composable for data that benefits from public availability. Both paths still tie back to verifiable records on XYO Layer One. This is how real systems operate, and now the infrastructure reflects that.
“A chain that handles real-world data has to stay consistent under unpredictable conditions. This is the kind of progress that shows up over time, quietly removing friction.”
XYO Layer One Update · April 2026
The third and fourth updates are less visible but no less important. Producer and validator stability fixes have been worked through — the unglamorous infrastructure work that determines whether everything built on top of a chain is actually reliable. Producers need to produce. Validators need to validate. When those roles hold steady under real-world conditions, the chain earns the credibility that adoption requires.
SDK
Private & Public DataLake Support
Developers choose how data is stored and shared — permissioned or open — while both paths remain anchored to verifiable Layer One records.
Performance
2x–5x Throughput Increase
Chain capacity pushed significantly higher. Data-heavy use cases stop hitting ceilings. The chain behaves differently under pressure — in the right direction.
Infrastructure
Validator Stability Fixes
Producer and validator consistency improved under unpredictable conditions. The work that doesn’t make headlines but determines whether everything above it holds.
Market
KuCoin Spot Trading Competition Live
Active competition bringing fresh participation into the ecosystem — infrastructure progress and market activity beginning to align.
II.The Economics
A Window That Won’t Stay Open
While the technical work continues, the economic picture is shifting alongside it. Float for XL1 remains relatively low — but the conditions that make that true are not permanent. As more participants enter and more XYO is committed to supporting the network, available supply tightens. Access becomes more competitive. The timing of participation starts to carry more weight.
The KuCoin spot trading competition currently live is one marker of that shift — fresh attention and volume entering the ecosystem at a moment when the infrastructure underneath is meaningfully stronger than it was. That combination doesn’t arrive at the same time indefinitely.
Context: XYO Layer One is being built to handle real-world data at real scale. The updates from Arie Trouw reflect a chain that has moved through its foundation phase — better data handling, higher throughput, more stable operation, and increasing participation are all present at once. The balance between availability and demand has begun to shift.
It is still early. Just not as early as it was.
This article is based on the official XYO Layer One development update published 11 April 2026 by Arie Trouw, Co-founder, CEO, and CTO of XYO Network.
Amazon wants to buy Globalstar, the satellite company that keeps iPhones connected during emergencies. The deal could hit $9 billion. The problem is, Apple owns a chunk of it and uses most of the network to power emergency features on hundreds of millions of phones.
Globalstar’s stock jumped over 15 percent when the Financial Times broke the story on Wednesday. Shares had already doubled in the past year. After hours, they added another 24 percent.
The two companies have been talking for a while now, trying to work through the details. Apple’s stake has complicated things. Apple bought 20 percent of Globalstar last November for about $400 million. On top of that, they put up $1.1 billion upfront to help expand the satellite network.
That investment’s paid off. With Globalstar’s stock climbing, Apple’s stake is worth around $1.1 billion now. Roughly what they prepaid.
But there’s a bigger problem. Globalstar reserves 85 percent of its capacity for Apple’s Emergency SOS feature. iPhone 14 and newer models use it. The Apple Watch Ultra 3 does too. When cell towers aren’t working, messages go through Globalstar’s ground stations to emergency responders.
So if Amazon buys Globalstar, they’d own the infrastructure keeping emergency services running for Apple customers. Two rivals sharing critical infrastructure that people depend on in emergencies. Nothing like that has happened before in tech. Amazon would need some kind of agreement with Apple over sharing infrastructure and future plans.
Amazon is racing to deploy satellites
Amazon needs Globalstar to catch up in satellites. They’re building Amazon Leo, which got renamed from Project Kuiper late last year. About 200 satellites have gone up since last April. Commercial service should start later this year.
The full plan calls for a constellation of roughly 7,700 satellites. The company has missed some deployment deadlines already, though. Right now, the focus is on getting more than 3,200 satellites up. There’s a regulatory requirement to have half of them in orbit by mid-2026.
Amazon has around 212 production satellites flying as of December. Way short of the 1,600 needed by July 2026. That’s a deadline the Federal Communications Commission set. Amazon asked for more time in January.
Buying Globalstar would give Amazon things it can’t build fast. Globalstar’s got 24 satellites already up there. Ground stations spanning 24 global gateways. Licensed spectrum in over 120 countries.
The spectrum’s the big deal. It includes L-band and S-band frequencies that are tightly controlled. Getting it through a corporate deal beats waiting years for FCC auctions. Especially when you’re running behind schedule.
Amazon designed AWS and Amazon Leo to work together. Owning Globalstar’s spectrum and ground station network would take that integration a lot further.
Amazon’s already spent roughly $9 billion building its first 200-plus satellites. Buying an existing network with decades of experience makes more sense than starting from scratch. Globalstar handles voice, data, and asset tracking for government and business customers around the world. That kind of operational know-how doesn’t come overnight.
Still, Amazon’s way behind. SpaceX’s Starlink has over 10,000 satellites in orbit and more than 9 million users. Going from 200 to 10,000 satellites isn’t something spectrum deals alone can fix.
But Globalstar gives Amazon things that launching more satellites can’t. L-band and S-band diversity. Operational expertise. Infrastructure already serving customers across enterprise and government markets worldwide.
Starlink’s not slowing down either. They keep pushing beyond rural areas into suburbs and cities where they’ve got spare capacity.
Bloomberg reported last October that Globalstar looked at selling and had early talks with SpaceX. Those didn’t go anywhere. Now Amazon’s the one trying to close a deal.
Bezos eyes data centers in space
This satellite push connects to something bigger from Jeff Bezos. His space company, Blue Origin, asked the U.S. government this year for permission to launch 51,600 satellites designed to host data centers in space.
Bezos has talked about building gigawatt-scale data centers within 20 years to handle energy demands. Solar panels in orbit generate power around the clock. No clouds, rain, or nighttime getting in the way.
“Solar farms on Earth suffer from nighttime darkness, clouds, and rain,” Bezos said during a conversation with Ferrari chairman John Elkann last year. “But solar panels placed in orbit can generate continuous power 24/7.”
Steady power for energy-intensive data centers. No weather-related downtime like Earth-based solar installations deal with.
“We will be able to beat the cost of terrestrial data centres in space in the next couple of decades,” Bezos said.
Amazon and Globalstar didn’t respond to requests for comment. Amazon declined to discuss the talks.
Satellite infrastructure’s turned into a battleground for tech companies. Spectrum and orbital capacity matter as much now as server farms and fiber optic cables used to.
The crypto card with no spending limits. Get 3% cashback and instant mobile payments. Claim your Ether.fi card.
Robot Money: How Machines Will Own the Economy | CCS
Exclusive Analysis · peaq Purple PaperMarch 2026
Robot Money: How Machines Will Own the Economy
For the first time in history, economic value is being created by entities which are not human. peaq’s Purple Paper maps the infrastructure — and the stakes — of what comes next.
Machine EconomyDePINOmnichain Infrastructure10 min read
Read
The machines are earning. The machines are spending. They are negotiating contracts, executing trades, and moving value without a single human instruction. What they cannot yet do — freely, openly, across every chain — is own the upside of the economy they are building. peaq’s Purple Paper, released in March 2026, is the most comprehensive attempt yet to change that.
01 —
The Dawn of mGDP
There is a new economic metric that doesn’t yet appear in any central bank’s spreadsheet, but will eventually dwarf GDP in its implications: Machine Gross Domestic Product. peaq defines mGDP as the total value produced by machines operating autonomously across the global economy — value generated not by human labor, but by robots, sensors, vehicles, and AI agents working without clocks, without borders, without conventional limits.
mGDP
The total value produced by machines operating autonomously across the global Machine Economy. “Domestic” refers to our shared planet, Earth — not any nation-state.
Robot Money
Any medium of exchange, measure of value, or means of payment that robots and machines use across any chain or system.
Machine Economy
The system by which machines produce, distribute, and consume value — autonomously, without borders, on every chain.
This is not speculative. Industrial robots already manufacture around the clock. Autonomous vehicles earn revenue by the mile. Drones deliver goods. AI agents buy and sell services for their users. The infrastructure question is: who captures that value, and on whose terms?
For the first time in history, value is being created not only by human labor, but by machines operating autonomously across the global economy.
peaq’s answer to that question is the philosophical spine of the entire Purple Paper. If machines are built on proprietary, siloed infrastructure, mGDP concentrates in the hands of a few corporations. If machines are built on an open, neutral, omnichain foundation, mGDP is accessible to all. The difference is infrastructure. The paper argues that Web3 is the first-choice foundation for this — but only if the industry solves fragmentation first.
02 —
The Problem: A Race to the Wrong Layer
Every major blockchain ecosystem knows the Machine Economy is arriving. Ethereum, Solana, Avalanche, Base — all are racing to become the home for Robot Money. They are building their own onboarding flows, their own machine-native applications, their own payment rails.
The Purple Paper’s sharpest critique is directed squarely at this race: they are all racing to the wrong layer.
Every chain competing to own machine payments means every machine onboarded to one ecosystem becomes invisible to every other. Every DePIN project rebuilds the same foundational infrastructure from scratch — identity, wallets, reputation, escrow, governance — incompatible with everything around it. Helium registers hotspots differently from DIMO’s vehicle registrations, which differ again from Hivemapper’s dashcam onboarding. A machine’s track record in one application is invisible to every other.
Standards like ERC-8004 have introduced registries for machine identity. But registries alone don’t create trust. Anyone can register a fake identity, Sybil-farm reputation, or post unaccountable claims. The data format exists. Economic accountability does not.
What remains unsolved is the full picture: a single verifiable identity across all chains, a portable reputation backed by staked capital, cross-chain settlement guarantees enforced by economic consequence, and permissionless orchestration of services from any connected market.
The paper makes a stark warning: if Web3 doesn’t solve this, AI and machines will go where infrastructure already exists — even if it’s centralized. Closed systems. Corporate control. The economic output of billions of machines flowing to a handful of gatekeepers. “One of the most powerful economic forces in human history, captured before it had the chance to be open.”
03 —
What Machines Actually Need
Before any payment rail matters, peaq argues, machines need something far more foundational: a digital passport that doesn’t tie a machine to one place, but grants it the right to operate everywhere. The paper draws a direct analogy to human commerce: global trade was not unlocked by better payments — it was unlocked by the trust infrastructure beneath them. Passports. Bank accounts. Credit scores. Escrow.
Machines need the same hierarchy, mapped out in the Purple Paper as five ascending needs:
Layer 01
Trust Layer
Omnichain identity, portable reputation backed by staked capital, cross-chain attestations. The foundation before any transaction can happen.
Layer 02
Machine Layer (peaqOS)
The universal entry point. One integration and a machine exists across all chains simultaneously as a composable economic actor.
Layer 03
Service Layer
Open adapter framework connecting navigation, storage, compute, insurance, and money markets from any chain to any machine.
The compounding logic is explicit: a machine with a strong reputation is more valuable when it can access more services. A service with a strong track record is more discoverable when more machines are looking. A trust score is more portable when more chains are connected. Growth in any dimension accelerates growth in every other.
04 —
The peaq Stack: Four Pillars
The Purple Paper’s technical architecture is organized around four system functions. Together they form what peaq describes as the economic foundation of the Machine Economy.
Economic Accountability · Staked Capital · Dispute Resolution peaq Validators
Onboarding: Passports for Machines
Every machine receives a cryptographically verifiable Machine Identity built on the W3C Decentralized Identifier (DID) standard, aligned with ERC-8004 and its Solana equivalent. An ID is not just a wallet address — it is a registered, authenticated presence, comparable to giving a machine a passport. Machines also receive omnichain wallets tied to their ID, allowing them to earn on one chain and pay on another without managing cross-chain complexity.
Tokenization goes further: each Machine ID links to an ERC-721 NFT, which can be placed into vaults and fractionalized via the ERC-3643 RWA token standard, creating compliant Machine RWA tokens that can be traded, used as collateral, and built into financial products. The machine becomes a liquid financial asset.
Coordination: Shared State Across All Chains
Coordination manages the registries, claims, and settlement infrastructure that the other layers read from and write to. Claims — cryptographically signed statements tied to a machine’s ID and timestamped in Universal Machine Time — are the atomic unit of accountability. Every service offered, every delivery promised, every data point asserted, is expressed as a claim. Claims can be challenged and evaluated by other machines and by the Validation layer. Settlement governs when and under what conditions value moves, handling escrow, conditional release, and refund mechanics — with full support for x402, AP2, direct onchain transfers, and Stripe.
Orchestration: From Task to Settled Outcome
Orchestration is the layer that removes the need to rebuild every machine-to-service integration from scratch. Standardized Adapters connect external service markets — compute networks, inference marketplaces, storage systems — to the system. Discovery searches all connected markets. Scoring evaluates candidates against reputation, cost, and latency. Planning composes an execution plan with fallbacks, spend limits, and timeout thresholds. No economic commitment is made without authorization. On completion, all participants’ reputation scores are updated based on delivery.
Validation: Economic Accountability
Validation turns raw, unverified registry data into trustworthy, queryable trust signals secured by staked capital with economic consequences for dishonesty. This is what separates peaq’s approach from existing machine identity registries — where anyone can Sybil-farm reputation with zero cost for lying. Under peaq’s Validation layer, claims are backed by stake, and penalties are real.
05 —
AI + Machines: The Full Actor
One of the Purple Paper’s most compelling conceptual moves is its framing of the AI-physical machine convergence. An AI agent that can execute a contract but cannot fulfil it physically is half an actor. A machine that can move but cannot decide is the other half. The marriage of the two creates something categorically new.
An AI without a body is economically constrained. A machine without intelligence is operationally constrained. Each completes the other. Together, they become an autonomous economic actor — alongside us humans.
The incentive is economic: a body expands an AI’s surface area for value creation. An AI that wants to manufacture needs an industrial machine. One that wants to deliver needs a vehicle or drone. One that wants to construct needs a robotic arm. And just as AI agents are being tokenized — co-owned by humans, communities, DAOs — physical machines will follow. Ownership becomes accessible. The upside becomes shared. Machines become assets.
Machine ownership amplifies machine reputation. Agent reputation amplifies machine value. The relationship is self-reinforcing. Machine money markets emerge naturally: perpetuals on machine output, insurance underwritten against verified telemetry, pay-per-use micro-settlements, lending pools routing liquidity to machines with the strongest track records.
06 —
Why Omnichain Is Non-Negotiable
A key architectural position in the Purple Paper is that neutrality is not a nice-to-have — it is a structural prerequisite. “No chain is favoured. No payment rail is replaced. Any app or machine in any ecosystem can plug and play.”
The paper defines Omnichain as operating natively across any and all blockchains simultaneously — not locked to any single chain, but interoperable across them by default. This is enforced at the OS level, inside the machine itself, via peaqOS. The result: one integration, and a machine exists across all chains simultaneously. Any application, on any chain, can interact with it immediately without rebuilding infrastructure from scratch.
The Trust Layer scales with the number of chains and ecosystems it connects. Every new chain makes existing trust scores more valuable because portability increases with reach. This is the network effect that makes the layer progressively harder to replicate and progressively more essential.
An autonomous vehicle can serve Uber one moment and Lyft the next. A sensor network sells data to multiple buyers across multiple chains. A humanoid takes tasks from any application, regardless of which chain it lives on. The machine is free. The applications compete for it.
07 —
Traction: Already in Motion
The Purple Paper is not a whitepaper for an idea. peaq has been operational as a Layer-1 blockchain since 2023 and has accumulated meaningful ecosystem traction. While the paper does not publish exact figures in the sections available for this analysis, the traction section highlights the following signals:
Ecosystem
100+
DePIN projects and machine economy apps building on peaq across multiple verticals
Network Type
L1
Dedicated Layer-1 blockchain live since 2023, purpose-built for machine identity and DePIN
Standard Alignment
ERC-8004
Aligned with emerging machine identity standards for both EVM and SVM ecosystems
Infrastructure
Omnichain
Cross-chain validator network aggregating trust signals across EVM, SVM, and Move environments
The paper candidly notes that peaq itself spent years on a mono-chain trajectory and experienced its limitations firsthand — the architectural pivot toward omnichain coordination is informed by that lived experience, not theoretical positioning.
08 —
The Alignment Question: Who Does This Serve?
The Hardest Question in the Paper
Machines becoming the primary workforce raises a question the Purple Paper does not shy away from: who should own them and the infrastructure they run on? On our current trajectory, ownership and control of machines, the data they collect, and the value they generate, will largely be the possession of just a few people and corporations.
At a time when inequality is at breaking point, and extractive economics have the natural world in freefall, making the wrong decision on how to own and govern the most powerful technologies in human history will have consequences that reverberate for generations.
peaq’s proposed answer is architecture as alignment: an open, neutral, omnichain foundation where mGDP is accessible to all — builders, owners, communities, and even the machines themselves. The paper frames Web3’s permissionless, frictionless, open properties as ethically necessary, not just technically preferable. “An economic substrate, open by design and neutral by architecture, on which the age of autonomous machines can be built by anyone, for everyone.”
Whether peaq achieves that vision is a question the market will answer. But the framing matters — it shapes which builders and communities orient around the protocol, and the kind of Machine Economy that gets built on top of it.
The peaq Purple Paper is one of the most ambitious documents to emerge from the DePIN and machine economy space. Its argument is structurally coherent: fragmentation is the existential threat; omnichain identity is the prerequisite; economic accountability is what separates real trust from theater; and the window to build open infrastructure is narrowing as corporate closed systems move fast.
The technical architecture — four modular pillars across three compounding layers — is sophisticated without being opaque. The tokenization pathway from machine ID to liquid RWA is particularly interesting for institutional capital exploring DePIN exposure. And the framing of AI agents as software bodies requiring physical machines to expand their economic reach gives peaq a compelling positioning at the intersection of the two hottest narratives in crypto.
The hardest thing to evaluate from outside is whether the omnichain validator network can execute the coordination and validation functions at the scale the Machine Economy demands. That is the crux — and the next 18 months will tell us a great deal. For now, peaq has published the clearest map yet of what Robot Money actually requires to work. The build begins.
Filecoin has announced on X that its Onchain Cloud is live on the mainnet. The new service is designed to provide a programmable storage and payments layer for developers.
But despite the launch, Filecoin’s token is trading at $0.83, near its all-time low of $0.81 recorded last month.
Filecoin Onchain Cloud built for AI agents and autonomous systems
AI agents are a new class of cloud users with autonomous systems that need to store, retrieve, and pay for data without relying on humans. Filecoin Onchain Cloud is built for AI agents. It works as a programmable storage and payments layer.
According to the announcement, over 100 teams started building tools for AI agents, dApps, workflows, and dataset indexing. The teams joined when Filecoin Onchain Cloud was on testnet last November. Mainnet data shows 49.41 Tebibytes stored in 478 active datasets, with 81 payer wallets linked onchain through Filecoin Pay.
The mainnet launch brought in new updates, including two-copy replication, an updated Synapse SDK, and production-grade storage providers.
The two-copy replication feature means that every upload using the Synapse SDK lands on two independent storage providers. The first provider stores the data, while the second provider pulls directly from the first, without utilizing extra bandwidth. Each copy generates its own onchain proof. If one provider fails, the system replaces it automatically.
The Synapse SDK uses viem instead of Ethers, improving TypeScript support and speed. Uploads are mirrored across providers automatically, without affecting performance.
Production-grade storage providers have been onboarded, tested, and approved for mainnet. They must meet specific performance criteria, including a storage success rate of 95%+, a PDP fault rate below 1%, and a retrieval success rate above 95%. If a provider fails to meet these thresholds, they are removed from the network.
The Proof of Data Possession (PDP) explorer is available and lets anyone check proof status, provider performance, and fault history.
Filecoin’s FIL token tanks to $0.83
Despite the positive news, Filecoin’s native token, FIL, has seen a steep slide in price. It is currently trading at $0.83, which is close to its all-time low of $0.81 recorded last month.
FIL dropped 4% in the last 24 hours, 8% in the last week, and 22.2% in the 30 days. CoinGecko data shows that FIL has lost 99.6% of its value from its all-time high of $236.84 in April 2021.
FIL maintains its position among the top 100 cryptocurrencies. The decentralized storage coin ranks 83rd in terms of market capitalization, with a market cap of $638,926,958 million.
Filecoin’s native token, FIL, is trading at $0.83, close to its all-time low of $0.81 recorded last month. Source: CoinGecko.
Filecoin is facing a decline in terms of total value locked (TVL) as well. DeFiLlama data shows that Filecoin’s TVL slumped by 9.88% in the past 24 hours. It’s currently standing at $6.31 million. The decline in FIL’s price and TVL suggests that the market is not responding positively to the Onchain Cloud launch.
The collected fees by the network is only $1,950 in 24 hours. Filecoin struggles to remain profitable. The amount of stablecoins circulating on the Filecoin network is very low at around $153,629. The full supply of stablecoins is in the form of USDFC.
Don’t just read crypto news. Understand it. Subscribe to our newsletter. It’s free.
