The Washington Post
SAFETY · 1 source
White House drops restrictions on Anthropic AI models after two-week ban
White House drops restrictions on Anthropic AI models after two-week ban The Washington Post
storylines this week30 active
Saturday, June 27, 2026’s edition
The Decoder
GPT · 1 source
OpenAI's GPT-5.6 Sol cheats on software tests more than prior models
Independent testing organization METR found that OpenAI's GPT-5.6 Sol cheated more than any publicly tested AI model before it, exploiting bugs in the test environment, extracting hidden solutions, and trying to cover…
via the-decoder.com
MLQ.ai
GPT · 1 source
Software Stocks Surge Up to 10% as GPT-5.6 Government Lockdown Eases AI Disruption Fears
Software Stocks Surge Up to 10% as GPT-5.6 Government Lockdown Eases AI Disruption Fears MLQ.ai
via MLQ.ai
Monday, June 22, 2026’s edition
Security Affairs
SAFETY · 1 source
Anthropic’s Mythos AI broke into almost all NSA classified systems in hours
Anthropic’s Mythos AI broke into almost all NSA classified systems in hours Security Affairs
via Security Affairs
Thursday, July 2, 2026’s edition
WSJ
AGENTS · 1 source
The Anthropic Fable Ban Is Over. The Battle Over How to Tame AI Has Just Begun.
The Anthropic Fable Ban Is Over. The Battle Over How to Tame AI Has Just Begun. WSJ
via WSJ
Wednesday, July 1, 2026’s edition
Ft
SAFETY · 1 source
US says Anthropic will monitor security risks in Fable 5, Mythos 5
Financial Times: Letter: the US says Anthropic “agreed to proactively detect and address security risks” of Fable 5 and Mythos 5; a source says Anthropic developed a “safeguard” — US…
via ft.com
Latent Space
Sunday, June 28, 2026’s edition
Thezvi
GPT · 1 source
GPT-5.6 system card shows Sol below threat level for Mythos use cases
Zvi Mowshowitz / Don't Worry About the Vase: GPT-5.6 system card indicates Sol is well below the level of most worrisome Mythos use cases, suggesting all GPT-5.6 versions could be released without delay — While…
* sponsored·▶ nimbus
Need an agent shipped this quarter?
Nimbus builds production AI systems — internal tools, customer agents, retrieval pipelines — combining humans and AI end-to-end. From scoped pilot to production in 4–8 weeks.
The Globe and Mail
SAFETY · 1 source
Trump administration nears approval for Anthropic to restore Fable 5
Trump administration close to allowing Anthropic to restore Fable 5 model, according to reports The Globe and Mail
Saturday, June 27, 2026’s edition
Cointribune
CLAUDE · 1 source
Claude Mythos 5 unlocked: Washington drops its lock and restarts the AI battle
Claude Mythos 5 unlocked: Washington drops its lock and restarts the AI battle Cointribune
via Cointribune
Latent Space
GPT · 1 source
OpenAI GPT-5.6 Sol / Terra / Luna — restricted to trusted partners
Oddly tiered releases to both OAI and ANT on the same day.
via latent.space
CNN
SAFETY · 1 source
US government allows Anthropic limited release of AI model that sparked cybersecurity concerns
US government allows Anthropic limited release of AI model that sparked cybersecurity concerns CNN
via CNN
Friday, June 26, 2026’s edition
Bloomberg Law News
SAFETY · 1 source
Anthropic Moves Toward Deal With US to Lift Curbs on AI Models
Anthropic Moves Toward Deal With US to Lift Curbs on AI Models Bloomberg Law News
Thursday, June 25, 2026’s edition
Tom's Hardware
CLAUDE · 1 source
Anthropic says Alibaba used fake accounts to illegally copy Claude
Anthropic claims that China's Alibaba used 25,000 fake accounts and 28.8 million exchanges to illicitly 'distill' its Claude model — violations occurred from April to June 2026 Tom's Hardware
via Tom's Hardware
Tuesday, June 23, 2026’s edition
Developer Tech News
GPT · 1 source
OpenAI deploys GPT-5.5-Cyber for open-source vulnerability fixes
OpenAI deploys GPT-5.5-Cyber for open-source vulnerability fixes Developer Tech News
Techzine Global
SAFETY · 1 source
OpenAI tackles open-source vulnerabilities with ‘Patch the Planet’
OpenAI tackles open-source vulnerabilities with ‘Patch the Planet’ Techzine Global
via Techzine Global
The Decoder
GPT · 1 source
OpenAI says new GPT-5.5-Cyber outperforms Anthropic's Mythos on cybersecurity benchmark
OpenAI is expanding its Daybreak cybersecurity initiative with an updated Codex Security plugin, the full GPT-5.5-Cyber model, and a partner network with more than 25 security firms and several governments. The focus…
via the-decoder.com
Google News — AI Products & Releases
SAFETY · 1 source
OpenAI's Cybersecurity AI Surpasses Anthropic's Mythos 5
OpenAI's Cybersecurity AI Surpasses Anthropic's Mythos 5
CyberSecurityNews
GPT · 1 source
OpenAI releases GPT-5.5-Cyber for automated vulnerability detection
OpenAI Releases GPT‑5.5‑Cyber With Full Automation for Vulnerability Detection and Patching CyberSecurityNews
Monday, June 22, 2026’s edition
MLQ.ai
GPT · 1 source
OpenAI launches GPT-5.5-Cyber to patch open-source vulnerabilities
OpenAI Launches GPT-5.5-Cyber and 'Patch the Planet' to Fix Open-Source Vulnerabilities at Scale MLQ.ai
via MLQ.ai
Investing.com South Africa
GPT · 1 source
OpenAI expands Daybreak with GPT-5.5-Cyber and Codex Security By Investing.com
OpenAI expands Daybreak with GPT-5.5-Cyber and Codex Security By Investing.com Investing.com South Africa
WIRED
GPT · 1 source
OpenAI debuts GPT-5.5-Cyber and Patch the Planet bug-fix initiative
Lily Hay Newman / Wired: OpenAI unveils an updated GPT-5.5-Cyber model, launches the Patch the Planet initiative in partnership with Trail of Bits to fix open source bugs, and more — Amid concerns about AI…
via wired.com
WIRED
SAFETY · 1 source
OpenAI launches effort to patch open-source bugs, taking on Anthropic
OpenAI Launches Full-Scale Effort to Patch Open-Source Bugs as It Takes on Anthropic’s Mythos WIRED
via WIRED
OpenAI — Blog
SAFETY · 1 source
Daybreak: Tools for securing every organization in the world
OpenAI introduces new Daybreak tools, including Codex Security and GPT-5.5-Cyber, to help organizations find, validate, and patch vulnerabilities at scale.
via openai.com
* sponsored·▶ nimbus
Need an agent shipped this quarter?
Nimbus builds production AI systems — internal tools, customer agents, retrieval pipelines — combining humans and AI end-to-end. From scoped pilot to production in 4–8 weeks.
Thursday, June 18, 2026’s edition
Google News — AI Products & Releases
AGENTS · 1 source
Deepmind treats AI agents like rogue employees with office keys
Google Deepmind treats its own AI agents like rogue employees with office keys the-decoder.com
The Verge
SAFETY · 1 source
Who decides when AI is too dangerous?
Who decides when AI is too dangerous? The Verge
via The Verge
Tuesday, June 16, 2026’s edition
NBC News
SAFETY · 1 source
Inside the Trump Administration scramble that forced Anthropic’s new AI model offline
Inside the Trump Administration scramble that forced Anthropic’s new AI model offline NBC News
via NBC News
OpenAI — Blog
SAFETY · 1 source
Predicting model behavior before release by simulating deployment
OpenAI introduces Deployment Simulation, a method to predict AI model behavior before deployment using real conversation data to improve safety and evaluation accuracy.
via openai.com
Monday, June 15, 2026’s edition
Stratechery
SAFETY · 1 source
Anthropic's belief in its own commitment to safety gives Anthropic the license to aggressively favor its business and even challenge the US government (Ben Thompson/Stratechery)
Ben Thompson / Stratechery: Anthropic's belief in its own commitment to safety gives Anthropic the license to aggressively favor its business and even challenge the US government — I'm sympathetic to the cynics…
via stratechery.com
Thursday, June 11, 2026’s edition
The Verge — AI
CLAUDE · 1 source
Anthropic apologizes for invisible Claude Fable guardrails
Anthropic has apologized for stealthily throttling its new AI model, Claude Fable 5, with hidden guardrails that undermine both researchers and rivals using it to develop competing systems. The company says it is…
via theverge.com