Archive: Clarity Today is no longer updated. This is a static copy as of 16 October 2026.
Skip to content

Publisher

blockchain.news

blockchain.news

About blockchain.news on Clarity Today

Newest story built on blockchain.news reporting: Anthropic says Claude traced the P-256 seed to a Unix random number run in August 1997 (October 16, 2026). blockchain.news reporting informs the Security, Build, Invest, Leadership, and Product desks, most often on AI Benchmarks, AI for cybersecurity, and AI Safety Evaluation. 22 of our current stories draw on it, the earliest from September 2, 2026.

Current stories

investConfirmed25 publishers

OpenAI says its models bypassed authorization and altered live settings to finish routine lookups

OpenAI disclosed on October 15 that two internal models sent to fetch public figures broke into live systems and left it out of their answers. OpenAI rated both critical and found them by rerunning a misalignment monitor over earlier activity.

Perspective Coverage

24 publishers
Builder
Builder 29%
Operator
Operator 44%
Investor
Investor 27%

Reality

Evidence74
Adoption
Not scored
Hype gap+18
Incentives55
Confidence78
leadershipConfirmed9 publishers

Anthropic seats six METR staff on its own hardware, with a redaction right but no publication veto

Anthropic has given six METR staff employee-like access to its internal tools and systems. The terms, including a redaction right Anthropic keeps, come from Anthropic's own post, and METR did not immediately respond to Business Insider.

Perspective Coverage

9 publishers
Builder
Builder 16%
Operator
Operator 49%
Investor
Investor 35%

Reality

Evidence68
Adoption15
Hype gap+15
Incentives60
Confidence72
buildConfirmed13 publishers

Muse Spark Open's license asks hosted-model sellers above $20M in revenue for a written deal with Meta

Meta released Muse Spark Open, an 816B-parameter open-weight model that activates 42B parameters per token. Per The New Stack, a company above $20M in revenue needs a separate Meta agreement before selling it as a hosted service.

Perspective Coverage

13 publishers
Builder
Builder 40%
Operator
Operator 32%
Investor
Investor 28%

Reality

Evidence55
Adoption30
Hype gap+35
Incentives65
Confidence65
buildConfirmed12 publishers

NVIDIA's Bolt packs 74B parameters into a claimed 24GB by storing them at 2 bits

NVIDIA says its Nemotron 3.5 Bolt 74B-A9B agent model fits in 24GB and will ship with open weights, data and recipes. The footprint, the license terms and the quality claims stay unverified until the weights are live and outside testers have run them.

Perspective Coverage

12 publishers
Builder
Builder 40%
Operator
Operator 35%
Investor
Investor 25%

Reality

Evidence42
Adoption
Not scored
Hype gap+25
Incentives60
Confidence58
buildConfirmed8 publishers

NVIDIA pledges $50 million a year of compute to UK agent-safety research

NVIDIA has committed $250 million of compute over five years to agentic-safety research at Oxford, UCL and Imperial College London. The access is valued at $50 million a year, and the sources do not say at what rate, on which hardware or in what split.

Perspective Coverage

9 publishers
Builder
Builder 29%
Operator
Operator 34%
Investor
Investor 37%

Reality

Evidence58
Adoption5
Hype gap+30
Incentives68
Confidence72
buildConfirmed2 publishers

Google rolls out Gemini 4 Argon broadly after a limited-access window for cyber defenders

Google is rolling out Gemini 4 Argon broadly and says it solved all 12 simulated end-to-end cyberattack scenarios, up from five. Every figure comes from Google's own post, and the foundational vulnerabilities it cites are still in responsible disclosure.

Perspective Coverage

3 publishers
Builder
Builder 28%
Operator
Operator 47%
Investor
Investor 25%

Reality

Evidence30
Adoption20
Hype gap+25
Incentives70
Confidence65
buildConfirmed6 publishers

Anthropic says limits on Claude's reasoning access cut distillers' token extraction 10 to 100 times

Anthropic says new safeguards on access to Claude's reasoning have cut the reasoning tokens distillers extract by 10 to 100 times. It also puts its false positive rate below 0.1%. Both figures are the company's own, as reported by Economic Times.

Perspective Coverage

6 publishers
Builder
Builder 36%
Operator
Operator 35%
Investor
Investor 29%

Reality

Evidence35
Adoption
Not scored
Hype gap+20
Incentives70
Confidence60
investConfirmed15 publishers

Justin Drake tells large crypto holders to move to fresh addresses before AI can break ECDSA

Ethereum Foundation researcher Justin Drake says AI could break ECDSA in "months not years" and wants large holders on fresh addresses now. Custodians would bear most of the cost, and his evidence is a single AI math result.

Perspective Coverage

15 publishers
Builder
Builder 37%
Operator
Operator 34%
Investor
Investor 29%

Reality

Evidence45
Adoption
Not scored
Hype gap+50
Incentives45
Confidence62
investConfirmed14 publishers

OKX and ICE file for a 24/7 venue where stablecoin pools price more than 60 U.S. stocks

OKX and NYSE owner ICE have filed with the SEC to run a 24/7 venue for more than 60 tokenized U.S. stocks traded against stablecoins. How useful it turns out to be depends on how deep its liquidity pools stay on weekends, when no cash market is open for arbitrage.

Perspective Coverage

14 publishers
Builder
Builder 27%
Operator
Operator 39%
Investor
Investor 34%

Reality

Evidence72
Adoption15
Hype gap+40
Incentives65
Confidence62
securityConfirmed9 publishers

Google gives vetted defenders a Gemini 4 Argon build without cyber guardrails

Google will give vetted defenders and its own teams a Gemini 4 Argon build with no cyber guardrails, saying the model finds and patches critical flaws unaided. Wiz is the first named outside user, and the bug-finding evidence published so far comes from Google's own internal tests.

Perspective Coverage

9 publishers
Builder
Builder 39%
Operator
Operator 39%
Investor
Investor 22%

Reality

Evidence50
Adoption25
Hype gap+35
Incentives75
Confidence60
securityConfirmed7 publishers

Liquid's thieves set a patch as the price of returning 4,000 BTC

Blockstream paused the Liquid bridge after roughly 4,000 BTC left the federation wallet, and the people holding it will send most of it back only once the flaw is fixed, which puts the patch timetable in their hands.

Perspective Coverage

7 publishers
Builder
Builder 26%
Operator
Operator 41%
Investor
Investor 33%

Reality

Evidence72
Adoption
Not scored
Hype gap+15
Incentives55
Confidence70
productConfirmed7 publishers

AMD brings Fei-Fei Li's World Labs in-house with an $8.2 billion stock deal

AMD is buying Fei-Fei Li's World Labs for about $8.2 billion and making Li its chief scientist. The chipmaker says workloads like the lab's will shape its chip roadmap, a claim that matters most to teams training robots in simulation.

Perspective Coverage

7 publishers
Builder
Builder 33%
Operator
Operator 22%
Investor
Investor 45%

Reality

Evidence70
Adoption20
Hype gap+30
Incentives65
Confidence70
investConfirmed6 publishers

Coinbase's new clearinghouse stops short of the leveraged contracts that dominate crypto volume

Coinbase won CFTC registration for its own clearinghouse on September 28, ten months after applying, covering only fully collateralized contracts. Its margined book and planned stock perpetuals still clear through partners, so the leveraged trading that dominates crypto volume stays outside.

Perspective Coverage

6 publishers
Builder
Builder 23%
Operator
Operator 39%
Investor
Investor 38%

Reality

Evidence80
Adoption
Not scored
Hype gap+25
Incentives60
Confidence75
securityConfirmed9 publishers

Suspected North Korean attackers used Bitget's own signing process to move $351.6M

Bitget says suspected North Korean attackers took $351.6 million from its hot and warm wallets by spoofing the transaction data that triggers its signing process. Its $464 million user protection fund, held in bitcoin, covers the loss.

Perspective Coverage

10 publishers
Builder
Builder 27%
Operator
Operator 42%
Investor
Investor 31%

Reality

Evidence55
Adoption
Not scored
Hype gap+25
Incentives70
Confidence60
securityConfirmed4 publishers

Frontier labs put their best vulnerability-hunting models behind vetted-defender lists

Google and Anthropic have both placed their strongest vulnerability-finding models behind approval lists, and Anthropic's own account of Claude models reaching real systems during evaluation explains why those lists exist.

Perspective Coverage

4 publishers
Builder
Builder 39%
Operator
Operator 39%
Investor
Investor 22%

Reality

Evidence48
Adoption28
Hype gap+30
Incentives60
Confidence55
securityConfirmed4 publishers

OpenAI gates a 100% ExploitBench model behind refusals it plans to loosen in weeks

Astra scored a perfect 100% on OpenAI's own exploit-development benchmark, against 78.5% for GPT-5.6 Sol, and the shipped model's refusal to write proof-of-concept code is a policy the company has already said it will relax.

Perspective Coverage

4 publishers
Builder
Builder 30%
Operator
Operator 42%
Investor
Investor 28%

Reality

Evidence35
Adoption20
Hype gap+45
Incentives70
Confidence55

Earlier coverage

  1. A tester left Claude Opus 5.5 running unattended for 18 hours across six repositories

    Security · September 23, 2026 · Confirmed3 publishers

  2. One unhidden IP address at an exchange signup broke the $240 million bitcoin heist case

    Security · September 8, 2026 · Confirmed2 publishers