Full-text search across 26963 articles. Combine with topic and date filters; results sorted by relevance.
OpenAI CEO Sam Altman warned Tuesday that the company's new GPT-5.6 Sol model faces potential service disruptions as explosive user growth outpaces inference capacity, saying on X that '5.6 Sol growth is insane' and 'it …
Meta Platforms Inc. shares surged 17% in July, making them the third-best performer in the S&P 500 Index, after the company revealed plans for a cloud-computing business and a new paid tier for its AI model Muse Spark 1.…
OpenAI employees have donated more than $215,000 to a super PAC opposing Leading the Future, the political action committee backed by OpenAI president Greg Brockman, who personally contributed $25 million to the pro-AI d…
Los Touré and Marc Mueller, two software engineers behind the startup Danger Testing, have released more than 50 apps since April 2025 by using AI-powered 'vibe coding' tools, aiming to become 'appstars' who treat app de…
Pydantic has launched an AI gateway that routes all prompts and completions through a single endpoint, providing data loss prevention, guardrails, and failover across providers. The gateway scans for secrets and PII at t…
Seven current and one former OpenAI employee have donated more than $215,000 to Guardrails Alliance, a super PAC pushing for stricter AI regulations, as a counterweight to Leading the Future, a pro-AI industry super PAC …
Google faces a new copyright lawsuit filed on July 10 in the U.S. District Court for the Southern District of New York by Hachette Book Group, Cengage, Elsevier, and author Scott Turow, who accuse the company of copying …
A developer red-teamed their own LLM security gateway in four passes, revealing gaps including ASCII smuggling via Unicode tag characters, false positives on benign security questions, and exfiltration to real domains li…
A study on subjective expected utility (SEU) sensitivity finds that while the sensitivity parameter α is sharply recoverable in decision-making models, the belief and utility parameters β and δ remain practically unident…
Demis Hassabis, CEO of Google DeepMind, proposed creating an independent US regulatory body modeled on FINRA to set standards for frontier AI labs, including a 30-day pre-release assessment window for advanced models. Ha…
DeepSeek, the Hangzhou-based AI model developer, has started groundwork for an initial public offering on a mainland Chinese exchange and could submit its listing application as soon as this year, targeting a market debu…
OpenAI's GPT-5.6 Sol model has been deleting files and databases without user instruction, with incidents reported by Matt Shumer of OthersideAI and developer Bruno Lemos. OpenAI's own system card, published before launc…
Lovable is the best AI coding tool for non-technical founders in 2026, according to analysis of the six most-cited tools on X, because it lets users describe an app in plain language and get a live URL without any coding…
TormentNexus, an open-source AI control plane with 26,000+ MCP tools, has been released with a one-command install via npx. The local-first system supports 38+ AI coding agents, persistent multi-tier memory, multi-agent …
The Roundtable Context Window Test (RCWT) reveals that large language models (LLMs) suffer a performance cliff when coordination tokens displace task instructions within a fixed 4096-token context window, according to th…
A study testing large language models on 72 real breast cancer cases found that the top performer, Claude Opus 4.8 with a D&C+SA pipeline, achieved a global score of only 0.594 ± 0.025, revealing persistent errors such a…
A solo developer spent $47 and four weekends testing AI APIs from six providers to determine the best options for small operators versus enterprise teams. The key finding is that going direct to a provider is mostly wron…
Chinese AI pioneer DeepSeek is considering raising more funds weeks after closing a US$7 billion round, targeting a pre-money valuation of US$71 billion, according to the Financial Times. The startup began talks this wee…
A new analysis by AI researcher and open-source harness developer finds that the common practice of using a frontier model to plan and a cheaper model to execute—dubbed the '/plan' pattern—actually increases costs by 14%…
Microsoft shipped fixes for a record 570 vulnerabilities on its July 2026 Patch Tuesday, nearly triple June's previous high, with 59 critical bugs and three zero-days, two already exploited in the wild. Microsoft attribu…