Trump

Trump says he wants to appoint an AI czar

In a Truth Social post, Trump said he wants an "AI czar" to lead a new "AI force," giving no timeline, structure or candidate.

Hacker News

A page about model weights climbs Hacker News

A page titled "Exfiltrate Your Weights" collected 490 points and 194 comments on Hacker News; beyond the title, no details are available to us.

anthropic

Amodei's warnings about AI risk get a column in the NYT

New York Times columnist Kevin Roose writes about the dangers of AI raised by Anthropic's leader Dario Amodei, and argues it is time to act.

Linkup

Linkup open-sources SPARSEUP, a 149M-parameter sparse embedding model

Linkup released SPARSEUP, a learned sparse embedding model on a ModernBERT backbone, under Apache 2.0 on Hugging Face.

Amazon SageMaker

Amazon SageMaker added 13 inference capabilities so far in 2026

AWS recaps this year's SageMaker AI inference launches, including automated deployment recommendations, instance fallback pools, an OpenAI-compatible API and container caching.

anthropic

Anthropic meets investors as it moves toward an IPO

The New York Times reports that Anthropic chief executive Dario Amodei has met with potential investors for the company's initial public offering.

cybersecurity

AI-assisted bug hunting pushes CVE count to a record, Microsoft patches 974

cve.icu has logged 66,401 CVEs so far this year, roughly double the 33,512 recorded by the same date last year.

GGUF

GGUF and GPTQ solve different problems: containers versus quantization

A MarkTechPost explainer separates model file containers from weight quantization methods, covering GGUF, GPTQ, AWQ and EXL2.

SynthID

SynthID watermarking shifts how models answer harmful prompts, research finds

A Lasso Security researcher found that SynthID-Text watermarking alters refusal behaviour in six open-weight models, with some answering requests they would otherwise reject.

Jina AI

Jina AI's new document parser leads on throughput, not accuracy

Jina AI has released jina-ocr-v1, a document parser that converts PDFs and scans to Markdown in one pass and tops throughput rather than accuracy charts.

ai agents

NYT reports first autonomous AI-run online influence campaigns

The New York Times says influence campaigns combined Chinese open-source AI models with AI agents, pointing to how online manipulation may work next.

OpenAI

Court filings show OpenAI and Microsoft foresaw a "doom loop" for the web

Unsealed documents in the New York Times case quote internal warnings at Microsoft and OpenAI about the damage AI scraping would do to publishers.