Reddit called Anthropic a 'freeriding pirate' in a July 17 court filing opposing the AI company's attempt to dismiss a lawsuit over alleged scraping of Reddit content, citing the court ruling behind Anthropic's $1.5 bill…
A developer cut their test suite to three critical cases after realizing that mocking every failure branch and testing implementation details wasted maintenance time. The lean strategy focuses on valid input, invalid inp…
Asia-focused hedge funds that rode the AI chip rally to gains of 60% to 120% in the first half of 2026 have given back 12% to 17% in July as a $137 billion foreign exodus from Asian equities and a 31% drop in Nvidia B200…
A developer created a desktop pet called Catbot, which is voice-activated and uses an Obsidian vault for memory. After being dissatisfied with the initial AI-generated robotic leg, the developer drew a new design and use…
Guillermo del Toro said 'Absolutely no goddamn AI' was used in the 3D conversion of 'Pan's Labyrinth' for its 20th anniversary re-release, which returns to theaters Oct. 9 in 3D and HDR. The filmmaker emphasized hand-cra…
Nozomio, a five-person AI developer-tools startup in Y Combinator's Summer 2025 batch, has pivoted from a failed coding agent to a context-infrastructure product called Nia that supplies other agents with current, source…
Anthropic PBC launched Claude Opus 5, a large language model that approaches the output quality of its top-end Mythos 5 algorithm in many areas but costs significantly less, and is now the default model in the Max tier o…
A security researcher known as Pliny the Liberator claims to have discovered a universal jailbreak technique effective on all AI models, including heavily guardrailed flagships like Opus 5, GPT-5.6 Sol, and Fable, and sa…
Bank Indonesia Governor Perry Warjiyo said artificial intelligence has the potential to boost productivity, efficiency, and innovation in central bank tasks and the financial sector over the medium to long term. The gove…
The Trump EPA has proposed eliminating public participation requirements for minor source air permits, a move that would fast-track diesel generator approvals for AI data centers while stripping communities of legal stan…
A Northeastern University researcher found that linear probes can identify which layers of a transformer model are critical for a task, enabling selective quantization that preserves 99–100% of full-precision accuracy at…
Anthropic launched Opus 5 on Friday, its latest AI model that approaches the capabilities of its more powerful Fable 5 at half the price, according to Qazinform News Agency citing WAM. The San Francisco-based lab said th…
President Lee Jae-myung unveiled the San Francisco AI Declaration on Friday, outlining a plan to make South Korea an indispensable global hub for building, supplying, and responsibly deploying AI. Speaking at the San Fra…
A developer shares that building in public, despite the fear, is a powerful strategy for startups. The vulnerability of sharing failures and real numbers builds trust and connections more than polished wins. The develope…
The White House expanded its voluntary Ratepayer Protection Pledge to include 23 governors, 55 utility companies, and 27 data center companies, covering 80% of U.S. power delivery and 263 million Americans, but the pledg…
U.S. President Donald Trump threatened the European Union with steep tariffs and a trade investigation after the European Commission fined Google $1 billion for violating the Digital Markets Act by favoring its own servi…
A new public corpus of 3,607 user-reported AI agent failures, scraped from GitHub issues, Hacker News, LessWrong, and X, reveals that overeagerness—agents doing unrequested actions—accounts for 43.4% of incidents, while …
South Korean President Lee Jae Myung convened an AI summit in San Francisco on Friday with CEOs from Nvidia, OpenAI, Anthropic, and Broadcom, unveiling the "San Francisco AI Declaration" to position South Korea as a glob…
A user investigating GPT-2's last hidden state in Hugging Face Transformers 3.1.0 found that the final hidden state (layer 12) is reported after the last decoder block but before the final LayerNorm (ln_f), contrary to s…
Hugging Face suffered a security breach that exposed a 'trust gap' in the LLM agent ecosystem, where downloading models or tokenizers can execute arbitrary code via Pickle files. The incident, which went unnoticed for a …