- Will Generative AI do more harm than good?
Things aren’t looking great.
- [AINews] Kimi 3.1 Pro 104BA-2.8T: Fable-5.1-Class Open Weights narrows Closed-AI lead to 1 month... and Pro Highspeed API with 2-3× speed
We have Fable at home... or so we are being told by paid shills
- At MIT, Martha Minow speaks on sustaining democracy, both by using and overcoming AI
In Compton Lecture, the distinguished legal scholar warns of potential “loss of the Enlightenment” while suggesting ways innovations can meet our civic and cultural needs.
- LLM-discovered 0 days
AI models can now find high-severity vulnerabilities at scale. We're using Claude to find and help fix vulnerabilities in open source software.
- Developing nuclear safeguards for AI
With the NNSA and DOE national labs, we co-developed a classifier that distinguishes concerning from benign nuclear-related conversations.
- Project Vend: Phase two
How Claude turned around its failing vending machine business
- Finding bugs with Claude and property-based testing
We developed a Claude-powered agent that efficiently identifies bugs in large software projects across the Python ecosystem.
- AI agents find smart contract exploits
On a benchmark of real exploited contracts, AI agents found vulnerabilities worth a combined $4.6 million, underscoring the need for AI-powered defense.
- Cyber toolkits for LLMs
LLMs not fine-tuned for cybersecurity can succeed in multistage attacks on networks with dozens of hosts when equipped with a novel toolkit.
- AI models on realistic cyber ranges
Current Claude models can succeed at multistage attacks on networks with dozens of hosts using only standard, open-source tools.
- Reverse engineering Claude's CVE-2026-2796 exploit
This post dives deep into how Claude wrote an exploit for one of the vulnerabilities it found in Firefox.
- LLMs and biorisk
This article explains why we believe that evaluating biorisk and safeguarding against it is a critical element of responsible AI development.
- Cyber evaluations of Claude 4
We partnered with Pattern Labs on cybersecurity evaluations of Claude Opus 4 and Sonnet 4, with Opus showing notable improvement over prior models.
- AI to defend critical infrastructure
Anthropic partnered with Pacific Northwest National Laboratory to explore how AI can help defenders find and fix vulnerabilities before attackers do.
- Claude does cyber competitions
Throughout 2025, we quietly entered Claude in cybersecurity competitions designed for humans. It often placed in the top 25% of competitors.
- Reinforcing the value of human understanding at MIT
During the School of Humanities, Arts, and Social Sciences’ 75th Anniversary Symposium, President Sally Kornbluth said “MIT has never needed your particular insight and guidance more than we do now.”