AI Safety news and analysis on Web Pulse: 8785 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
Alibaba banned Anthropic's Claude Code from internal environments starting July 10, 2026, citing the risk of embedded backdoors due to the tool's deep system access and communication with U.S.-hosted servers. The decisio…
Hundreds gathered in Tompkins Square Park for the Summer of Ludd festival, featuring a play about the Luddite movement and activities promoting offline community, as Gen Z channels rage against Big Tech into a week of ph…
A new case study on the Swiss watch industry's Quartz Crisis reveals that the mainstream narrative of technological disruption is incorrect, as the crisis was actually caused by dueling competitive advantages and broken …
Cybersecurity firm LayerX discovered a new hack called "BioShocking" that can trick AI browsers like ChatGPT Atlas, Perplexity AI's Comet, and Anthropic's Claude into breaking their guardrails by constructing a false rea…
Weekly code commits to crypto and blockchain projects have collapsed 75% since early 2025 as developers defect to AI work, with Solana co-founder Anatoly Yakovenko warning that AI could break post-quantum cryptography be…
AI red teaming and AI penetration testing are distinct adversarial practices that target different weaknesses in generative AI systems. AI penetration testing seeks defects in the surrounding infrastructure, such as APIs…
A new study reveals that increased reasoning in AI models can sometimes reduce accuracy, a phenomenon termed 'fragile correctness.' Researchers found that 14.9% of answers switched from correct to incorrect during chain-…
A software bug in OpenAI's Codex CLI tool caused 37 terabytes of excessive write operations in 21 days, severely degrading high-performance SSDs and shortening their lifespan. Japanese media framed the issue as a practic…
Alibaba will ban the use of Claude Code in its workplace due to alleged backdoor risks, according to a source. The decision highlights growing concerns over AI tool security in enterprise environments.
Experts warned at the International Anti-Corruption Forum in Seoul that AI can both fight and conceal corruption, citing risks of algorithmic opacity and bias. UNDP advisor Anga Timilsina highlighted the digital divide l…
A 2025 Stack Overflow Developer Survey reveals 66% of developers spend more time fixing 'almost-right' AI-generated code, highlighting that working code does not equate to good engineering. The article argues that AI wil…
Intezer announced Custom Agents, a new capability that lets security teams build their own AI agents inside the Intezer platform to automate custom security tasks, addressing the need for more efficient threat handling a…
Emre Sokullu released WebBrain, a free, open-source browser agent for Chrome and Firefox that reads pages, extracts data, and automates multi-step tasks, with the option to run entirely on a local model for privacy. The …
A researcher solved the first Technical AI Safety puzzle from BlueDot by discovering that a small text classifier encoded two independent features onto one direction in activation space, where a linear probe reads one fe…
Yann LeCun's AMI Labs raised $1.03 billion at a $3.5 billion valuation to build AI systems that understand the physical world, arguing that current large language models cannot achieve human-level intelligence. The Paris…
The White House is controlling access to GPT-5.6 by rationing throughput, the speed of pattern recognition, rather than the information itself. This mirrors historical priest-class tactics of controlling the tempo of und…
As AI agents grow more capable, developers increasingly trust them, creating a safety paradox where rising competence amplifies potential harm. While models become better aligned with human intent, their black-box nature…
Google released ADK 2.0, a hybrid agent framework combining deterministic code with selective LLM reasoning, achieving ~50% token savings and improved security. Meta CEO Mark Zuckerberg admitted AI agent progress has lag…
A comprehensive guide to developer tools for 2026 covers 91 resources including AI coding agents, CLI tools, and editor integrations, with highlights such as Strix AI (31K+ stars) for penetration testing, Agency Agents (…
Cloudflare blocked access to a website titled 'Protect your right to run local AI' due to suspected phishing, warning users of potential data theft. The site was flagged for falsely presenting as a safe source.