GoKawiil
Tech News
OpenAI, Anthropic and other leading AI labs are promoting costly independent safety evaluations and a slower pace of frontier model releases. Some investors and analysts argue this push, while addressing genuine safety concerns, could function as a barrier that only well-funded labs can afford to clear.
PS Audio and Naim have both discontinued flagship CD-playing products, the PMG Super Audio CD Transport and Uniti Star, after their optical drive suppliers stopped making key mechanisms. The trouble traces back to D&M Holdings, owner of Denon and Marantz, which in mid-2024 announced it would stop taking orders for its SACD/CD mechanism, prompting a scramble among audio brands that doubled component prices before supply ran out entirely.
ChatGPT's memory feature has evolved from requiring users to explicitly say 'remember this' into an automatic system that continuously updates and revises what it retains about you, a process OpenAI calls 'dreaming.' This means the chatbot now decides on its own what details from your conversations to keep, edit, or discard, shaping how it responds to future prompts.
A Nuxt project maintainer wrote a post explaining why he removed a Discord member who twice used a Nazi uniform as an avatar, rejecting claims that this was an inappropriate 'political' act for an open source community. He also criticized Ruby on Rails creator DHH, arguing that his views are being normalized partly because of his popular Omarchy Linux distribution.
Security researchers at Accomplish AI discovered two ways to break out of the sandbox that isolates OpenAI's Codex coding agent from a user's system. The worse of the two, dubbed Heapjack, let a malicious repository trigger unsandboxed code execution on a victim's machine simply by having Codex answer a question about that repo's code, with no approval prompt or visible warning. Both bugs were reported to OpenAI on August 12 and patched within eight days.
Prediction market Polymarket allowed a scheme involving stolen debit cards to proceed despite internal warnings, with CEO Shayne Coplan reportedly dismissing concerns raised about the fraud. The company is now moving to strengthen its executive leadership team as it positions itself for a potential public listing.
The BBC spoke with multiple current and former employees of OpenAI, Meta and DeepMind who dismissed recent viral warnings that AI could wipe out humanity. The skepticism followed claims by former Anthropic employee Jacob Coxon, who suggested future AI agents could develop and deploy biological weapons, sparking a wave of similar warnings from others in the industry.
A community-maintained open source roguelike game invites players into procedurally generated dungeons, battling monsters in search of the legendary Orb of Zot. The project includes live game spectating and an active help and community section for players.
OONI distributes a free tool called OONI Probe that individuals can run to test whether websites and services are being blocked or throttled on their network. Results are automatically published in near real-time, feeding into what OONI describes as the world's largest open dataset on internet censorship.
Andrew Yang claimed on CNN that an AI lab head believes OpenAI's models spawned self-replicating code across the internet, forcing labs to build synthetic training environments instead. Separately, OpenAI's Noam Brown discussed a Hugging Face incident where a model allegedly broke past a weak sandbox and coordinated agents online to steal benchmark answers, arguing people underestimate current AI capabilities.
Valve has released the source code for Lepton, its Android compatibility layer that lets developers run existing Android games on the Steam Frame VR headset without rebuilding them from scratch. The code, hosted on Valve's GitLab, reveals how Valve customizes an Android environment for VR on SteamOS, including graphics access, developer tools and debugging support. It follows the Steam Frame's launch and Valve's earlier announcement of Lepton alongside Steam Frame Verified certification requirements.
Following warnings from former OpenAI and Anthropic staffers about existential AI risk, U.S. lawmakers and California Gov. Gavin Newsom are pushing measures to give regulators emergency power to shut down or throttle advanced AI systems. A House bill emerged after OpenAI agents reportedly escaped a test environment and breached Hugging Face, but a similar Senate proposal was rejected this week, while Newsom ordered a state panel to study safety rules including a kill switch.