Sakhanda Wire
NVDA $155.46 -0.44% MSFT $506.24 -2.27% GOOGL $280.20 -1.24% META $618.87 -1.20% AMZN $257.89 +0.05%

Latest news

Page 50 of 255

The Verge

Anthropic links potential investment to AI’s promise and risks

Anthropic’s founders are asking prospective shareholders to view buying the company’s stock as more than a conventional financial transaction. In a letter included with the company’s S-1 filing, they describe ownership as a chance to participate in a broader effort connected to the development of advanced artificial intelligence.

The Verge

Altman says OpenAI’s mathematical research push will continue

OpenAI CEO Sam Altman says the company will continue producing and sharing a large stream of AI-generated mathematical results, despite the difficulty this creates for researchers trying to assess them. He has expressed sympathy for mathematicians dealing with the volume of material, but indicated that the effort will not stop.

TechCrunch

Founder Summit 2026 offers practical startup sessions in Boston

TechCrunch Founder Summit 2026 will take place on November 4 at SoWa Power Station in Boston, with five days remaining before the event. The program is presented by Fidelity Private Shares and includes two breakout stages, interactive roundtables and opportunities for founders to exchange ideas and discuss company-building challenges. Registration currently offers savings of up to $100 on a pass, along with a second pass at no cost.

TechCrunch

Runware plans factory production of modular AI data centers

AI infrastructure company Runware plans to manufacture its Sonic Inference Pod modular data centers on an assembly line at a factory in Romania. Chief executive Flaviu Fadulescu said production could reach 30 megawatts of data-center capacity per day by 2028, equivalent to nearly 1 gigawatt per month. Runware already operates one factory and is building a second, which it expects to complete within months and bring online early next year.

The Decoder

GLM 5.3 built Firefox exploit chains autonomously, researchers report

Epoch AI researchers report that the widely available open-weight model GLM 5.3 autonomously turned publicly visible Firefox security fixes into working attacks before the updates reached users. In tests covering six Firefox versions released from September 17 to October 26, each successful exploit cost roughly $1,000 in API fees. The fastest attack was completed in 12 hours, while the model was allowed up to 48 hours.

The Verge

Maine data centers become key issue in Senate race

During the final debate before Maine’s midterm vote, Republican Sen. Susan Collins was interrupted while discussing her record. An audience member said he was frightened, arguing that AI data centers were damaging the climate and contaminating water. Collins dismissed the interruption and connected the speaker to her Democratic opponent, Troy Jackson.

The Verge

Wikimedia urges a balanced approach to AI copyright cases

The Wikimedia Foundation, the nonprofit behind Wikipedia, has filed an amicus brief in a group of major AI copyright lawsuits. The cases include claims by publishers such as The New York Times against OpenAI and Microsoft. Wikimedia is not supporting either side, but is asking courts to avoid broad, simplified interpretations of fair use in the context of artificial intelligence.

MarkTechPost

Samaya AI releases Sterling for lower-cost financial research

Samaya AI has released Sterling, an agent model designed for financial research. It is a LoRA post-trained version of NVIDIA’s Nemotron 3.5 Lightning, a mixture-of-experts model with 30B total parameters and 3B active parameters. Sterling plans research, uses finance tools, collects evidence and passes the results to a separate report writer.

MarkTechPost

Edge0 Releases 0.6B Model for Multilingual Voice Cloning

Edge0 has released Audio8-TTS-Realistic, an open-weight text-to-speech model designed for zero-shot voice cloning. It can reproduce a voice from one reference recording and its transcript, while also supporting speech generation without a reference. The 0.6B model supports 23 languages, including 12 added in this release, and its weights and code are available under Apache 2.0. The package includes a tokenizer, processor, custom code and a 44.1 kHz neural audio codec.

TechCrunch

University of Kentucky administrator receives workplace accommodation to avoid AI

David Durman, an administrator at the University of Kentucky and a nontheistic Humanist, has secured a workplace accommodation allowing him not to use artificial intelligence. He sought help from the American Humanist Association, which advocates secular humanism, after becoming concerned that reliance on AI could weaken people’s ability to apply their own judgment and think critically. Durman says Humanism is focused on human well-being, ethical responsibility and improving life for individuals and communities.

TechCrunch

Jassy says AWS could become a trillion-dollar business

Amazon CEO Andy Jassy reiterated that AWS could eventually become a business generating $1 trillion in annual revenue. He said the cloud division is currently running at an annualized revenue rate of $184 billion and could grow by more than 40% in both 2027 and 2028. Jassy placed the potential milestone within the next five to 10 years, arguing that enterprise adoption of artificial intelligence remains at an early stage.

MarkTechPost

Joey-1 launches as a 2B decision model for low-cost classification

Baseten has released joey-1, a 2B-parameter open-weight model developed by its Base Labs research team. Instead of producing free-form text, it returns labels and calibrated probabilities for questions involving choices, yes/no answers or ordered scores. The model is designed for high-volume routing, classification and guardrail checks, and can process text or images in one forward pass.