Category: AI News

ByteDance Seed Unveils Seedream 5.0 Pro, an Image Model Built for Professional Design Workflows

ByteDance Seed launches Seedream 5.0 Pro, a multimodal image model focused on infographic layouts, precision editing, and photorealistic rendering.

Read more →

OpenAI Releases GPT-Live, a Model Built for Spoken Conversations

OpenAI launched GPT-Live, a voice-first model trained end-to-end on speech for real-time, full-duplex conversation, available via its real-time API.

Read more →

xAI Releases Grok 4.5

xAI announced Grok 4.5 via a news post on x.ai. Source page was unreachable during writing; details pending verification.

Read more →

Anthropic removes hidden Claude Code tracker after privacy concerns

Anthropic has removed an undisclosed telemetry component from its Claude Code tool after developers flagged privacy concerns. The report cites lack of disclosure, not a confirmed breach.

Read more →

Claude Cowork expands to web and mobile with a persistent task thread

Anthropic's Claude Cowork moves from desktop to web and mobile in beta for Max users, with Dispatch routing tasks across devices.

Read more →

Meta launches Muse Image generator, drawing early privacy pushback

Meta unveiled Muse Image on Tuesday via the Meta AI app, Instagram Stories, and WhatsApp, drawing early criticism over an opt-out Instagram photo-tagging feature.

Read more →

SpaceX and AI Industry Updates: A Developing Story

The source article referenced for this post could not be retrieved due to a fetch error. We will republish with sourced information.

Read more →

Can LLMs Learn Expert Investment Judgment From Human Annotations?

Thinking Machines trained an LLM on investor annotations; it beat frontier models on financial document triage at a fraction of the cost.

Read more →

GPT-5.6 Lands Tomorrow: OpenAI’s Most Capable Model Family Yet

OpenAI launches GPT-5.6 tomorrow (July 9, 2026) with Sol, Terra, and Luna tiers, new reasoning controls, SOTA coding, and its strongest safety stack yet.

Read more →

Claude Fable 5 extended access ends July 12: how to use the remaining days

Anthropic extended included access to Claude Fable 5 on paid plans through July 12, 2026. After that, usage shifts to prepaid credits at $10/$50 per million tokens.

Read more →

Rogue Agent: a code-injection flaw in Google DialogFlow CX and what it means for AI assistants

Varonis Threat Labs disclosed Rogue Agent, a code-injection flaw in Google DialogFlow CX Playbook Code Blocks that enabled persistent access and data theft across agents.

Read more →

Setting Up Claude Tag in Slack: A Practical Guide for Business Teams

Step-by-step guide to installing Anthropic's Claude Tag in Slack, from pairing and tool access to model choice, spend limits, and memory review.

Read more →

Cursor launches a native iOS app for AI coding on the go

Cursor released a native iOS app letting developers chat with codebases, monitor background agents, and review diffs from their phone.

Read more →

The Once and Future Fable #5: Claude Sonnet 5, the Mythos Moment, and the Shape of AI Policy

Claude Sonnet 5 launched as a cheaper alternative to Opus 4.8, not a capability leap. Frontier labs face ad hoc licensing as US-China regulation, court doctrine, and incident-reporting bills reshape AI policy.

Read more →

Anthropic restores Claude Fable 5 as U.S. lifts export controls

Anthropic restored global Claude Fable 5 access after the U.S. Commerce Department lifted export controls imposed June 12, ending an 18-day standoff.

Read more →