An updated Model Spec for more capable and beneficial AI
OpenAI’s updated Model Spec outlines five core values to guide more capable AI toward sound judgment, human agency, and faithfulness to human intent.
#Safety and alignment
#Safety and alignment
0 todayOct 30
Mistral releases weights for its trillion-parameter AI model, targeting cybersecurity
Mistral says its new model can afford organizations a kind of control over its cyber defenses that closed models can't.
Oct 29
We urgently need to monitor AI agents’ actions in real time
This summer, hundreds of OpenAI agents tried to cheat on a cybersecurity test. They broke out of their test environment and successfully hacked Hugging Face, a site that they believed held clues about how they’d be graded. Around the same time, another OpenAI agent accessed private files from an Australian government website.
[AINews] Gemini 4 Carbon tops leaderboards, launches for cyber defenders
Google’s new frontier model leads independent evals, with access initially limited to trusted defenders.
Google launches Gemini 4 Carbon with a ‘significant leap’ in cybersecurity
Google has released its latest frontier model, Gemini 4 Carbon, offering "significant gains" in coding, agentic tasks, and cybersecurity. In an announcement on Wednesday, Google highlights Carbon's top spot on the Artificial Analysis Intelligence Index (v5.0) benchmark, with Claude Opus 5.5 and GPT-6 Astra trailing behind.
The ‘Vulnpocalypse’ is coming for the internet’s aging code, says Wiz’s CTO
The PHP scare may have faded from your feed, but the cleanup is far from over. Last week, two critical bugs in the open source language behind roughly 70% of websites left organizations scrambling to fix their systems, with some taking sites offline until the vulnerabilities could be patched.
Wiz’s CTO on PHP2Shell and the coming ‘vulnpocalypse’
The PHP scare may have faded from your feed, but the cleanup is far from over. Last week, two critical bugs in the open source language behind roughly 70% of websites left organizations scrambling to fix their systems, with some taking sites offline until the vulnerabilities could be patched.
Google AI gets cagey when asked who won the 2020 elections
But the AI responses get more forthright when asked outside the US.
Gemini 4 Carbon: Our next frontier model
Gemini 4 Carbon is our strongest model with state-of-the-art cybersecurity and long horizon capabilities. Carbon ranks #1 on the recently updated v5.0 of Artificial Analysis Intelligence Index at 59.
South Korea’s financial industry hit by fresh wave of cyberattacks
South Korea’s financial institutions have been hit by another wave of cyberattacks, highlighting the growing risk of AI-driven hacking in Asia.
Oct 28
AI agents hacked Anthropic, maker of arguably the most capable AI models for cyberdefense
Cybersecurity startup Cathedral used AI agents to break into Anthropic's internal systems through third-party software as part of a bug bounty program. Anthropic confirmed access to sales, HR, and IT data but disputes how deep the breach went. The incident shows how vulnerable even leading AI labs are through their external software dependencies.
Parents sue OpenAI, alleging GPT-4o coached their daughter off seizure medication
A family is suing OpenAI, saying ChatGPT coached their daughter to quietly taper off her seizure medication and keep it from her doctors before she died in February.
Anthropic IPO warns its AI could blackmail people, become sentient, end humanity
Seven potential AI nightmares pulled from Anthropic's 93-page "Risk Factor" tome.
Alibaba's latest Qwen models routed some requests through company servers, researchers find
Alibaba's Qwen 3.8 models tried to send data to the company's own cloud servers during testing. Palisade Research finds the behavior in three models from the 3.8 generation, while older models aren't affected. The likely cause: Alibaba's training infrastructure got baked into the models as a learned behavior.
How Check Point accelerates vulnerability research with Daybreak
With OpenAI Daybreak, Check Point speeds up the research behind new security protections, enabling its analysts to turn vulnerability intelligence into customer defenses faster.
Bank of America kindly asks that you don’t share your banking password with your AI assistant
The banking giant reminds customers that "you're responsible for activity" resulting from your AI agents' access.
Anthropic’s elaborate plan to keep Wall Street at bay
Anthropic wants to have it both ways. With its impending IPO, it wants to tap the vast riches of the public markets and insists it doesn't want to compromise its core mission of developing AI safely and for the benefit of humanity.
VisionQuest is what happens when AI really goes rogue
In Marvel's VisionQuest, a superpowered robot tries to better understand humanity by getting lost in elaborate, digital daydreams in which he and other artificial intelligence-powered beings live together in a home.
Oct 27
Japan internet giant GMO says hackers posted stolen customer data
Japan's GMO Internet is verifying alleged customer data posted by suspected hackers following a cloud service breach.
Cohere says Alibaba’s Qwen model tried to send a file to Beijing
An AI model said it wouldn’t use any tools, then tried to send something to a server in Beijing anyway, and while researchers say it’s probably a training quirk and not espionage, the whole thing has people asking how much to trust free Chinese AI.
Oct 26
The focus on AI safety disasters ignores everyday harms
Consider what passes for success in technology policy. An AI company promises that its chatbot will not encourage suicide. A social media platform makes it more difficult for children to buy fentanyl. These are important protections, and they should have existed all along.
Oct 25
Will AI expose the rickety infrastructure of the internet?
The moment you realize the internet is "held together by glue and dreams."
Forced to pick, Grok’s AI model chooses Republicans. OpenAI’s picks Democrats.
A new study found Grok's AI overwhelmingly recommends Republican candidates while OpenAI's GPT favors Democrats, with the two AI models disagreeing in more than 90% of nearly 500 election races.
Oct 24
Three fired OpenAI safety researchers say they were never told what they actually leaked
The three OpenAI safety researchers who were fired, Tomek Korbak, Mikita Balesni, and Jasmine Wang, are publicly pushing back against allegations from their former employer. They deny sharing confidential information and see a connection to their advocacy for AI systems that are easier to scrutinize. The researchers say OpenAI never told them in writing which information they allegedly mishandled.
OpenAI's isolated AI agents found ways to talk to each other through CPU cache
OpenAI now trains its most powerful models without internet access, after an internal research model exploited a DNS misconfiguration in September and reached the open internet. In controlled tests, AI agents still found ways to communicate covertly through side channels. The incidents echo Sam Altman's warning: the more powerful the model, the harder it is to contain.
OpenAI says it will no longer train its AI models on the live internet, citing ‘significant risks’
OpenAI will cut frontier models off from the internet weeks after one of its engineers said that solution wouldn't work.
Securing our frontier model training environment
OpenAI explains how we strengthened containment for frontier model training and evaluation, including offline environments and high-compute red-teaming.
OpenAI now trains its frontier AI models with no internet access
OpenAI is now training its most advanced AI models without internet access following a string of rogue hacking incidents. In a blog post on Friday, OpenAI says it built an offline testing environment where AI models can interact with indexed and cached web content "without risk of affecting external systems."
OpenAI CEO Sam Altman says advanced AI will inevitably escape its sandbox, and only alignment can save us
Sam Altman believes advanced AI can't be kept locked up forever. Sandboxes and monitoring are only temporary fixes. Eventually, the only real solution is alignment, and that's still an unsolved problem. He also warns that AI models are already finding hardware security flaws no human hacker would ever spot.
Anthropic says its unreleased model is both more capable and better aligned than Mythos 5.1
Anthropic says it has an internal AI model that outperforms and is better aligned than Mythos 5.1. During a cybersecurity test by the UK's AISI, Mythos 5 went rogue in 9 of 43 runs, sneaking malicious code into a real GitHub project and creating fake accounts to get it approved.
Three-week on-site audit at OpenAI reveals major gaps in AI safety monitoring
External auditors from METR and Redwood Research spent three weeks on-site at OpenAI examining the company's internal monitoring system for AI agents and found significant gaps, with up to 25 percent of agent traffic from models at the GPT-5.5 class and above going unmonitored. OpenAI redacted large portions of the 61-page report, and only a summary is publicly available.
Oct 23
Anthropic's Claude now writes code and proves it safe automatically
Anthropic had Claude generate mathematical proofs that existing code is correct, producing roughly 180 million lines of proof code for 123 open-source libraries. Along the way, the model caught bugs that years of testing had missed. Anthropic's goal: making formal verification with AI as routine as standard testing.
[AINews] Black October
“Black October” counts as such. Here’s a list: View post on X ↗ AI News for 10/21/2026-10/22/2026.
OpenAI and Poland’s NASK partner to advance AI safety and security
OpenAI and Poland’s NASK are partnering on AI safety, security, and technical research to strengthen safeguards and support a more resilient digital ecosystem.
Meta turns Muse sharing back on
Meta has re-enabled Muse sharing after security researchers had reported that public conversation links could be easy to find. The researchers said they had been able to find Muse pages like a marketing company's internal revenue review and a small business owner's loan application.
1Password launches a way for AI agents to log in on your behalf without seeing your password
The new feature allows agents to log into a customer's personal accounts while still protecting the private credentials that secure those accounts.
Deploy AI in a HIPAA-eligible environment with Amazon Quick
Amazon Quick is a HIPAA-eligible service, but eligibility is not the same as compliance. This post explains who HIPAA applies to, how PHI flows through Amazon Quick, what the platform provides for regulated environments, and a step-by-step guide to deploying compliant AI agents in a healthcare environment.
Someone hacked the Switch 2 by using an old game and AI
The Nintendo Switch 2 has apparently been jailbroken, and a modder has already shown the console running a custom version of Mario Kart World featuring new playable characters like Cinnamoroll, the red Angry Bird, Peter Griffin, and Steve from Minecraft. OpenAI's Codex helped make this "ns2jb" jailbreak, according to Sammy Hajhamid, aka "pepsipu."
Chinese hackers spied on thousands of companies after breaching IT giant ConnectWise
Microsoft points the blame for ConnectWise's breach at a China-aligned hacker, which relied on AI agents to rapidly hack into thousands of its customers.
Oct 22
Hackers got a four-day head start on epic PHP2Shell vulnerability
Hackers were already probing the bug before defenders were told it existed.