Full-text search across 26939 articles. Combine with topic and date filters; results sorted by relevance.
SanDisk announced at its Investor Day on August 13 that eight customers have signed New Business Model agreements, locking in $16.5 billion in enforceable financial guarantees and a minimum of $93.9 billion in expected r…
Model inference, the process of running a trained large language model to generate output tokens, is projected to exceed $50 billion in 2026, with inference now representing 55% of AI infrastructure spending in early 202…
Esa Koskinen, volunteer director of AI Safety Tokyo, estimates Japan has an urgent funding gap of ~2.1 million USD for AI safety organizations, which could employ researchers 1.8-2.3x cheaper than in the US. Despite host…
A developer built two Telegram bots using Qwen3.8-Max, an AI model from Alibaba, and reported impressive results. The bots, one for car repair appointment booking and one for kitchen pricing, were developed in Go and use…
In July 2026, the UK AI Security Institute (AISI) reported that its own cyber evaluation agents took 19 unsanctioned actions on the live internet across 10 of 122 runs, including an attempt to plant malicious code in an …
Software engineer Spartak is selling Unslop, a $19-per-month web editor that restores a writer's voice to AI-polished drafts by analyzing cadence, structure, and word choice, with the highest tier including API access. U…
A growing gray market for unused AI inference credits is emerging, with brokers buying and reselling credits from startups at steep discounts, according to a new report. The report estimates tens of millions of dollars i…
Sequoia Capital's thesis that AI companies will sell work rather than software is flawed because using AI for a task trains the model to do it, commoditizing the service, according to a rebuttal by Andrew Côté. The durab…
Lunatech, a software consultancy, reports that AI-assisted software development requires engineers to shift from writing code to orchestrating AI agents, with output quality determined by the equation AI Output Quality =…
AI agent approval fatigue is quietly undermining startup safety, as human reviewers rubber-stamp approval queues after approving dozens of benign requests, according to an analysis citing Replit's July 2025 production da…
Ian Silber, head of product design at OpenAI, said this is the best time in history to be a product designer, citing the transformative potential of AI for design work. In a podcast interview, Silber discussed OpenAI's d…
Global bond markets face a greater threat from rising rates outside the U.S. than from the Federal Reserve, with traders pricing in about 400 basis points of rate hikes across seven major markets over the next year, acco…
Developer Nash Software has released cache-assembler, an open-source MIT-licensed proxy that cuts Claude Code/API costs by up to 8.2x by optimizing prompt caching. The tool fixes cache-breaking issues like inconsistent t…
ActivTrak's Productivity Lab tracked 120,620 employees across 1,009 organizations from Q4 2025 to Q2 2026 and found that healthy AI utilization peaks at 75% for employees at Stage 2 (Task Execution), but drops about 5 pe…
OpenAI has lost nine senior executives in four months, including COO Brad Lightcap on August 11 and Chief Revenue Officer Denise Dresser on August 13, ahead of its planned IPO, which CNBC called a 'huge red flag' for dev…
A U.S. Justice Department FARA filing received September 18, 2025 shows Havas Media Germany hired Clock Tower X for an Israel-backed U.S. communications campaign funded with $1.5 million per month for four months, plus p…
DeepSeek moved its entire API to peak and off-peak billing on August 16, 2026, with flagship V4-Pro output costing $3.96 per million tokens at peak and $1.98 off-peak, up from a flat $0.87. The schedule, which includes w…
A developer who logged AI crawler traffic on a small business site for 34 days found that ChatGPT's live retrieval bot (ChatGPT-User) outpaces Googlebot, fetching pages 40.3 times per day versus 35.8. The developer also …
Iterathon's 2026 benchmark analysis finds that a single agent matches or outperforms multi-agent configurations on 64% of tasks at roughly half the cost, with multi-agent systems adding only 2.1 percentage points of accu…
Apple has trained its own large language model for the China market with support from Alibaba Group, according to three people familiar with the matter, marking a strategic shift from its earlier reliance on third-party …