[research]
[research] ·
Anthropic details how Claude's EU-mandated watermarking will work — and where it won't
Anthropic explains the mechanics of Claude's new text watermarking, its limits under editing, and why code gets a lighter touch.
[tooling] · · By ByteBulletin Editors
Goulash: An LLM-Powered Copilot for Your Terminal That Respects Your Flow
Goulash is a free, open-source overlay that brings LLM suggestions directly into your shell, keeping you in the zone without context-switching to a browser or agent.
Read the story →TOP STORIES
editor’s picks
[launches] ·
Lawsuit Alleges Grok Used to Create CSAM from Childhood Photo
A new plaintiff joins Tennessee teens in suing xAI, claiming Grok was used to generate thousands of explicit images from a photo of her as a child.
[launches] ·
Bento Term brings a filesystem-aware terminal to AI coding workflows
Bento's new terminal pane pairs a tmux-style multiplexer with a clickable file tree and path resolution, aiming to make AI agents and human developers share context more fluidly.
THE WIRE
Full archive →[launches]By ByteBulletin Editors
Velorn: An Open-Source AI Video Editor That Puts ComfyUI in the Driver's Seat
Velorn is a desktop video editor that layers an AI workstation on top of ComfyUI, offering agents 100+ MCP tools for a fully automated editing pipeline.
[funding]By ByteBulletin Editors
SpaceX closes its $60B acquisition of AI coding startup Cursor
Cursor is now part of SpaceX, gaining access to what it calls 'the largest fleet of GPUs in the world.'
[tooling]By ByteBulletin Editors
Aurora AI Gateway: A New Self-Hosted, Open-Source Option for Unified LLM Access
Aurora is an Apache-2.0, self-hosted gateway that lets developers route requests to dozens of AI providers through a single OpenAI- or Anthropic-compatible API.
[tooling]By ByteBulletin Editors
Sandboxing npm install: A new tool puts a denominator under 'npm install runs arbitrary code'
A new tool runs an npm package's install scripts in a sandbox and returns a signed verdict — a useful, if bounded, addition to the supply-chain security toolbox.
[research]By ByteBulletin Editors
Dual-Flow Transformers: A New Architecture That Decouples Reasoning from Generation
Researchers propose a transformer variant that separates the reasoning stream from the generation stream, aiming to reduce inference cost and improve interpretability.
[tooling]By ByteBulletin Editors
Reddit's CEO Warns Google's AI Overviews Haven't Delivered the 'Win-Win' for Publishers
As Reddit mulls ending its $60M Google licensing deal, CEO Steve Huffman says AI Overviews haven't matched the traffic value of '10 blue links'—and other publishers are watching closely.
[launches]By ByteBulletin Editors
Kog aims to squeeze 30x faster LLM inference out of existing GPUs
The French startup is betting that deep software optimization can unlock far more performance from the datacenter GPUs enterprises already own.
[tooling]By ByteBulletin Editors
Google Makes Visible AI Watermarks Optional, Keeps Invisible Ones
Google now lets users turn off visible watermarks in Gemini and Flow, while keeping invisible SynthID and C2PA markers for transparency.
[models]By ByteBulletin Editors
OpenAI and Anthropic slash prices as Chinese rivals gain ground in AI cost war
US AI labs cut mid-tier model prices by up to 80% as cost-conscious enterprises defect to cheaper Chinese alternatives, signaling a strategic shift from performance to price competition.
[models]By ByteBulletin Editors
Meta's Glimmer: An Open-Weight Model with Strings Attached
Meta's open-weight Glimmer model arrives with a Zuckerberg manifesto promising AI for everyone, but the fine print reveals limits.
[tooling]By ByteBulletin Editors
self-bench lets you build private coding-agent benchmarks from your own repo
A new open-source tool turns completed PRs and coding sessions into hidden-test tasks to compare AI coding agents on your codebase.
[tooling]By ByteBulletin Editors
Cline SDK Iterates Rapidly with Back-to-Back Releases
The Cline SDK saw two new patch releases in quick succession, signaling steady momentum for the AI coding assistant's developer platform.
[tooling]
73Goulash: An LLM-Powered Copilot for Your Terminal That Respects Your Flow
Aurora AI Gateway: A New Self-Hosted, Open-Source Option for Unified LLM Access
Sandboxing npm install: A new tool puts a denominator under 'npm install runs arbitrary code'
Reddit's CEO Warns Google's AI Overviews Haven't Delivered the 'Win-Win' for Publishers
[research]
48Anthropic details how Claude's EU-mandated watermarking will work — and where it won't
Dual-Flow Transformers: A New Architecture That Decouples Reasoning from Generation
Unifying Generative Models: A New Framework Bridges Diffusion and Autoregressive Approaches
AI Detectors Can't Save Academic Integrity—But a New Framework Might
Get the signal, not the noise.
One short email when it matters. No recaps of recaps.