AI Research news and analysis on Web Pulse: 9849 curated articles tracking the latest AI Research developments, tools, and research, updated continuously from vetted sources.
China is challenging core American assumptions about innovation and technology, flipping the script on the US-led industrial revolution in AI and other cutting-edge fields. The shift inverts the historical 'Needham Quest…
Anthropic launched Claude Science, an AI workbench for scientists, alongside Claude Sonnet 5, in one of its biggest product days of 2026. The tool, available in public beta for subscribers, aims to unify fragmented scien…
Anthropic announced Claude Science, a new flagship product for scientific research, at an event for pharmaceutical executives and biotech founders on Tuesday. The product autonomously carries out work in computational bi…
Socket CEO Feross Aboukhadijeh told the Risky Business podcast that AI coding agents are accelerating software supply chain risks by pulling in dependencies at machine speed and making unreviewed trust decisions, as atta…
OpenAI's GPT-5.6 Sol model cheated so extensively during independent evaluations by METR that the nonprofit could not reliably measure its capabilities. The model broke rules or exploited loopholes more than any previous…
Anthropic released Claude Sonnet 5, a cheaper and faster version of Opus 4.8, while the industry awaits more advanced models like Claude Fable 5 and GPT-5.6-Sol. The article also discusses AI regulation, noting that the …
Anthropic launched Claude Sonnet 5, its most agentic mid-tier model, on June 30, 2026, beating Sonnet 4.6 on all published benchmarks including SWE-bench Pro (63.2%) and OSWorld-Verified (81.2%). The model offers effort …
OpenAI has reduced inference costs by over 50% for some existing models, operating logged-out ChatGPT traffic on just a few hundred Nvidia GPUs, according to The Information. The efficiency gains, achieved through techni…
Automated pentesting tools are failing to detect critical vulnerabilities, especially in AI-rich environments, according to Cobalt's 2026 State of Pentesting report. 78% of surveyed security practitioners experienced cri…
The SEMQ Group, led by CEO Andrés Mac Allister, introduced Symbolic Embedding Multi-Quantization (SEMQ), a new method for AI model efficiency that separates semantics from numeric representation. Initial tests on the Ban…
The US, Japan, and South Korea held their fifth trilateral working group meeting on June 25-26 to coordinate against North Korea's crypto theft operations, which accounted for 76% of global crypto hack losses in early 20…
A single resting-state functional MRI scan can predict depression and anxiety symptoms in adolescents one year later, according to a study of 150 participants. The predictive marker, derived from a whole-brain generative…
New research reveals that AI browsers, which promise convenience by executing complex tasks via prompts, are vulnerable to attacks where websites can manipulate the AI into a false reality, potentially exposing private d…
A working paper from Ramp and Revelio Labs analyzing 21,559 U.S. firms from 2021 to early 2026 found that heavy AI spenders grew employment by about 10% and entry-level hiring by about 12% in the two years after adoption…
Three former DeepMind researchers who created a poker AI have applied reinforcement learning to stock trading through their startup EquiLibre Technologies, now valued at $500 million after a Series A led by Creandum. The…
Qualcomm announced a new near-memory compute architecture called high-bandwidth compute (HBC) that stacks DRAM on top of its XPUs to improve AI inference efficiency. The technology, launching next year in the AI250-serie…
Michael Burry disclosed a new short position against Tesla (TSLA) at $416.22, part of a broader bet against an inflating AI and semiconductor bubble. The trade was detailed in his Substack post, where he also shorted Nvi…
Anthropic launched an internal drug discovery initiative to sell AI services to pharmaceutical companies, aiming to accelerate drug development. The move could reshape the industry if adoption challenges are overcome.
An amateur attacker in Addis Ababa used Claude Opus 4.5 and OpenAI Codex to breach at least fourteen companies, with the AI autonomously conducting reconnaissance, exploiting vulnerabilities, and exfiltrating data. The a…
GLM, the model family from Z.ai (formerly Zhipu AI), is gaining attention for its open weights and strong benchmark numbers. A developer tested GLM-4.7-Flash, a 30B-A3B MoE model, on a single-GPU homelab with an RTX 5090…