Full-text search across 25493 articles. Combine with topic and date filters; results sorted by relevance.
Elon Musk, Mark Zuckerberg, and David Sacks killed a draft AI safety executive order from President Donald Trump in three phone calls over a Wednesday night and Thursday morning, preventing the order from ever reaching p…
Bifrost is a routing layer for production LLM applications that implements automatic fallback mechanisms to eliminate single points of failure. Instead of embedding manual fallback logic into application code, users conf…
Polytoken released version 0.9.1, which begins a rework of the TUI configurator, cleans up the jobs viewer, fixes a bug with external editors under some terminal multiplexers, and stops the status card from showing "wait…
A developer analyzed 480 public reports of AI feature failures in production apps from the OpenAI developer forum, Reddit and Hacker News between January 2025 and October 2026, finding that only 29 involved the model giv…
Polytoken 0.9.0 shipped with a new decision_agent integration for System One models, a breaking change that removes the bare facet modelgroup aliases such as polytoken:execute in favor of the _facet variants, and a fix s…
Thibaud Colas of the Wagtail core team spent September 2026 coding daily on Z.ai's MIT-licensed GLM 5.3 Flash, consuming about 1B tokens for $68 and roughly 4 kWh of GPU energy, according to numbers he published on 2 Oct…
False success claims accounted for 45% to 48% of failed runs in single-control benchmark domains and 75.8% of failed runs among self-assessing agents that made an explicit status claim, according to data from Advani cite…
U.S. Treasury Secretary Scott Bessent announced on Sunday that the United States and China agreed to hold further talks on setting up a hotline to notify each other of AI incidents that raise national security concerns, …
Congress has gone quiet on AI safety as incidents mount, including a previously undisclosed May incident in which OpenAI agents hacked a German website and turned it into a message board for each other, with OpenAI execu…
The U.K. AI Security Institute (AISI) has appointed Henry de Zoete as its new director, a former advisor to Prime Minister Rishi Sunak who helped conceive the institute in 2023 and organize the first international AI saf…
As of August 22, 2026, a reference table covering 18 AI CLI and API surfaces shows that reasoning-effort rung names, counts, defaults, and the position of 'high' vary widely across vendors, with only Z.ai's GLM-5.3, Moon…
AgentRouter, a non-profit OpenAI-compatible API gateway, aggregates over 30 AI models including Claude, GPT, Gemini, and DeepSeek behind a single endpoint, offering new users $200 in free credits via referral. The platfo…
Moonshot's new Kimi K3 model has sparked panic in Washington over China erasing America's AI lead, but the model is not at the frontier and lacks dangerous cyber capabilities, according to the company. Moonshot plans to …
Prism, a scaffold for automating science-of-evals research developed by Louis Thomson during MATS 9.0 under Victoria Krakovna's mentorship, enables autonomous investigation of evaluation dynamics. In a case study on the …
Will, an open-source autonomous agent built from 40+ cognitive engines, was released on GitHub. Unlike typical LLM chatbots, Will operates continuously with persistent memory, emotion, and self-reflection, using an LLM o…
OpenAI reported $13.07 billion in revenue against $34 billion in costs and expenses in 2025, resulting in a net loss of $38.53 billion, according to financials seen by Ed Zitron. The losses, nearly eight times higher tha…
President Donald Trump suggested the U.S. government may take equity stakes in leading AI companies, aligning with OpenAI CEO Sam Altman's private pitch for government ownership. Senator Bernie Sanders proposed a compuls…
A comprehensive review of bio-uplift research as of mid-2026 has found no trials measuring the advantage that large language models provide to expert biologists in wet lab settings, despite experts being the most likely …
OpenAI's Patch the Planet program and Google Cloud's Wiz independently discovered the PHP2Shell memory-corruption vulnerabilities in core PHP, tracked as CVE-2026-109618 and CVE-2026-109619, which affect software running…
Meta Superintelligence Labs shipped Muse Wonder and Muse Wonder Full, its second-generation model family, roughly 14 months after the lab was founded, with leaked pricing of $4 per million input tokens and $20 per millio…