personality clashes sent anthropic's models offline
axios reports internal friction and export control issues at anthropic, with key staff meeting the commerce department amid model safety concerns.
ailatest summaries
plain 200-word summaries of important ai and data science news, updated twice daily.
measuring trust between ai agents
researchers propose a behavioral measure of trust between ai agents using costly verification in a survival game, finding that larger models reduce verification with reliable teammates but differ in recovery after failures.
aigpu workflow for hypersonic flow emulators
a fully gpu-based method speeds up building neural emulators for hypersonic flows with physics-aware refinement and uncertainty quantification.
aistanford students protest google ceo over military ai ties
about 200 stanford graduates walked out and booed sundar pichai during his commencement speech, protesting google's cloud and ai contracts with israel's military and u.s. immigration enforcement.
aius anthropic ban was never about ai jailbreak
the us government forced anthropic to pull its latest ai models offline using an export control directive, raising concerns about political interference in ai.
aimeta adds ai search to facebook using public posts
meta launches ai mode on facebook to answer questions by summarizing public posts, groups, and reels, alongside new editing tools and photo presets.
aisalesforce buys fin for $3.6b to boost ai agents
salesforce acquires ai customer service platform fin for $3.6 billion to enhance its agentforce platform with advanced agent technology.
aisarvam becomes india's newest ai unicorn with $234m round
bengaluru-based sarvam raised $234 million at a $1.5 billion valuation, led by hcltech, to build sovereign ai models and enterprise applications for indian languages and sectors.
aicybersecurity experts protest us ban on anthropic models
dozens of cybersecurity veterans signed an open letter urging the us government to lift export controls on anthropic's fable and mythos models, arguing the ban harms defenders.
ainewcore raises $66m to manage ai agent identities
cybersecurity startup newcore emerges from stealth with $66 million in funding to authenticate, govern, and control ai agents at scale.
airecursive diffusion training converges to a smoothed limit
even with perfect learning, early stopping in diffusion models causes geometric drift toward a unique limiting distribution that acts as a low-pass filter on the data.
aisatellite uses ai to find things on its own in orbit
a satellite ran a vision-language model in space for the first time, letting it answer natural language queries about earth imagery without human help.
ailocalized multidirectional correction for refusal suppression in moe models
a new method called lomc suppresses refusals in mixture-of-experts models using compact, support-gated edits that preserve general performance.
aiuser portrait planning adapts dialogue strategies in real time
a new framework uses user portraits and nested rollout to let large language models dynamically tailor dialogue policies without offline training.
aioptimal transport recovers geometric shifts for domain adaptation
a study shows optimal transport with p-norm cost can recover rotation, translation, or scaling between source and target domains in 2d, enabling linear regression adaptation with scarce target data.
aiai layoffs surge as wealth gap widens
tech companies cite ai for record layoffs while insiders amass huge wealth, fueling skepticism and public anger.
aidecision trees unified via bregman divergences
a new framework extends decision trees to any bregman divergence loss, generalizing classic cart and enabling flexible, interpretable models for diverse data types.
aitransformer policy for open shop scheduling via deep reinforcement learning
a transformer-based scheduling policy trained with deep reinforcement learning solves open shop scheduling problems and scales to large instances without retraining.
aimobile npu framework speeds up diffusion llm inference
llada.cpp is the first npu-aware framework for accelerating diffusion large language models on smartphones, using multi-block speculative decoding and other techniques.
aiwhy ai hasn't replaced software engineers
ai speeds up coding but software engineering bottlenecks remain in deciding what to build, verifying outputs, and deep human understanding.
ai