Only Wed, October 21, 2026 · All dates →
USC names AI school after $200M Stevens giftUSC celebrated Mark and Mary Stevens' $200M gift by naming its new school the USC Mark and Mary Stevens School of Computing and Artificial Intelligence. One of the largest gifts in USC history, it will fund researcher recruitment and accelerate discoveries across health, medicine, business and other fields.
Blackstone AI head says buildout debt fears are overblownBlackstone AI head Jas Khaira said at TechCrunch Disrupt that market fears over AI buildout debt are overblown. With $1.3T in assets, Blackstone launched San Francisco AI hub BXN1 this year and committed $5B to Crux AI, a neocloud built with Google, and backed chip startup Etched. Khaira said Blackstone is the largest owner of compute outside hyperscalers.
October Capital Launches to Focus on AI Infrastructure ServicesOctober Capital, a new firm formed by tech-focused private equity shop Siris, has launched to focus on services supporting AI infrastructure, physical security, and financial infrastructure. CEO Frank Baker said the firm will target service opportunities in these areas.
NASA Seeks Data and AI ChiefNASA is seeking a Data and AI Chief to lead the agency's data and AI strategy, reflecting growing emphasis on AI leadership roles in government agencies.
HHS Plans Sole-Source Anthropic Claude ContractThe U.S. Department of Health and Human Services (HHS) plans a sole-source contract with Anthropic for Claude ahead of an enterprise AI BPA, signaling federal agencies' accelerating adoption of enterprise AI tools.
OpenAI Works with Distribution Partners on Consistent SafeguardsOpenAI outlined how it collaborates with distribution partners to maintain consistent safeguards, support responsible access, and clarify expectations in its Usage Policies. The move highlights uniform standards as access expands.
IBM Q3 sales up 5% to $17.1B, CEO cites AI tailwindIBM's third-quarter revenue rose 5% to $17.1 billion, beating estimates, with software sales expanding and EPS of $2.95. CEO Arvind Krishna called AI a tailwind for the business.
SAP Cloud Growth Beats Forecast, Signaling Software ResilienceSAP's cloud products grew faster than expected, indicating the AI boom has not eroded demand for the German software maker's offerings. The results are seen as a sign of resilience in enterprise software demand.
Ex-Harvard AI adviser joins Anthropic oversight trustHarvard physics and astronomy professor Christopher W. Stubbs joined Anthropic's Long-Term Benefit Trust, which can appoint a majority of the board and advises on AI risks. Less than three weeks earlier he stepped down as the Faculty of Arts and Sciences' senior adviser on generative AI, where he helped secure access to models including Claude.
Bose CEO Lila Snyder discusses Skullcandy acquisition and AI roleBose has acquired Skullcandy, and CEO Lila Snyder told Bloomberg the deal's implications for both companies, how Bose aims to reach younger users with wired headphones, and AI's role in future products. The deal amount was not disclosed.
Meta plans new non-voting stock to fund AI spendingMeta Platforms plans to introduce a new class of non-voting stock that could fund its heavy AI spending, according to Semafor. The move would let Meta raise capital without diluting founder control; size and timing were not disclosed.
Apple unveils new human-crafted Apple Music brandApple revealed a new Liquid Glass-style brand identity for Apple Music. The author sees the design language as a passing trend, but argues that a human-crafted brand identity carries extra weight in an era of AI-generated content.
Walmart launches Agent Gateway for AI agent shoppingWalmart launched Agent Gateway, an MCP-based server from Anthropic's protocol, letting approved AI agents search inventory and buy for users with spending and transaction controls across Walmart and Sam's Club. It contrasts with Amazon's blocking of AI agents; Walmart already partners with Meta's Muse and integrates ChatGPT and Gemini.
Mistral releases open-weight multimodal model Large 4Mistral AI released Mistral Large 4 in public preview, an open-weight, general-purpose multimodal model with a fine-grained Mixture-of-Experts architecture: 52B active parameters, 1.05T total parameters, and a 1.6B vision encoder.
Harvard mathematicians split over OpenAI math breakthroughAfter OpenAI said it solved the roughly 90-year-old Navier-Stokes problem, Harvard mathematicians disagreed: some worry AI proofs lack citations and explanations and are hard to evaluate, while others say accuracy matters more than origin. A September-formed advisory group on math and AI aims to guide how AI research is released.
Google staff resign over military AI contractsBBC spoke to several Google DeepMind staff who resigned or are leaving over the company's AI contracts with the US and Israeli militaries; London research engineer Sofia Liguori resigned this week. Google said contracts go through compliance reviews and that it has channels for employee concerns.
Anthropic launches proactive agent Claude DuetAnthropic launched Claude Duet, a proactive agent powered by Opus 5.5 that keeps working while users are away and waits for approval when needed. Available to Pro and Max plans in the US, it works with more than 6,000 connectors and plugins.
AI Will Become A Commodity Over Time: NamburiMadhu Namburi, managing director at General Catalyst, said AI will become a commodity over time. He discussed AI's impact on dealmaking and investments on Bloomberg.
Anthropic publishes public benefit charter with five pillarsAnthropic, a Public Benefit Corporation, published a charter and report committing to balance shareholder interests with public benefit. It outlines five pillars: solving AI safety, universal safety practices backed by law, democracy and global stability, prosperity and flourishing, and shared benefits.
NVIDIA launches Kumo Relational foundation modelNVIDIA Kumo Relational is a pre-trained relational foundation model that predicts directly on enterprise databases without task-specific training or feature engineering. Across 41 tasks on 15 databases it beats supervised relational deep learning and expert-tuned gradient-boosted trees, and on SAP SALT it tops AutoGluon ensembles by 8% in mean reciprocal rank. It scales to 500B+ rows and is available via an open-source library or NVIDIA NIM.
Judge rejects Trump claim EPA can dismiss data center lawsuitsA US judge rejected Trump's claim that the EPA can unilaterally dismiss Clean Air Act lawsuits, so xAI (now SpaceXAI) still faces a suit over alleged pollution from its Memphis Colossus data center. The NAACP seeks best-available pollution controls and penalties up to $124,426 per day per violation.
G.M. Hopes to Design New Cars in as Little as 2 Years Using A.I.General Motors aims to use artificial intelligence to cut new car design time to as little as two years. The move could significantly speed up vehicle development and reduce costs, though details on implementation and investment were not disclosed.
Train Microduck to walk in Isaac Lab 3.0 and deploy to real robotPollen Robotics and Hugging Face released Microduck, an affordable open-source bipedal robot, with a tutorial for training walking policies in GPU-parallel simulation using NVIDIA Isaac Lab and Newton. Actuator modeling and domain randomization prepare policies for real hardware; policies export to ONNX for deployment, can be shared on the Hugging Face Hub, and can compete in Microduck Arena.
Florida lawyer referred for discipline over 'AI slop' filingA Florida lawyer was referred for possible discipline after filing court documents that judges called 'AI slop.' The case highlights growing judicial scrutiny of low-quality or inaccurate AI-generated submissions.
Websites scramble to fix software flaw that could put much of internet at riskWebsites are rushing to patch a software flaw that could put much of the internet at risk. The vulnerability, if exploited, could cause widespread security problems. Details of the flaw and affected software were not disclosed.
Data center boom sparks debate over benefits, costs and risksU.S. data center investment is surging, boosting the economy while drawing opposition. Global capacity tripled from 2015 to 2024 and accelerated after ChatGPT's launch. Supporters call data centers the infrastructure backbone for AI; critics warn of a debt-financed bust, exaggerated local jobs and tax benefits, energy and environmental costs borne by communities, and risks from AI itself.
Specializing SLM vocabularies can cut inference cost by 77%A preview of the VocabX paper shows that rebuilding the Qwen3-4B vocabulary for a single domain and output format can greatly cut inference cost. In a Portuguese essay-grading system, the method designed the vocabulary around token cost rather than frequency; with the same base model, data and training recipe, measured inference cost fell 77%, while shrinking the roughly 151,669-entry vocabulary scored at each generated token.
Blacksmith raises $66M Series C at $1B valuationBlacksmith, a platform for building and testing AI-written code, raised a $66M Series C led by Sequoia at a $1B valuation. ARR grew 17x and it has run over 90 million CI jobs; it was valued at $550M just two months ago.
Report flags legal risks of AI-generated drug evidenceA Crowell & Moring client alert says probabilistic AI models in drug discovery can produce plausible-but-wrong outputs that affect regulatory submissions, trial integrity and legal liability. It lists five risks: hallucinations, unrepresentative training data, opaque reasoning, input sensitivity and adversarial manipulation of training data, and urges experimental validation, human oversight and preparation for FDA and EU guidance.
Poll: 57% of Americans say AI is bad for societyAn ABC/Washington Post/Ipsos poll finds 57% of Americans say AI is bad for society versus 42% good, and about two-thirds oppose new US data centers, with stronger opposition locally. Among those under 30, 65% view AI negatively, while Republicans are more evenly split.
The A.I. Daily