Agents
Microsoft releases Scope, an open source test tool for coding agents
Scope operates one task with different coding agents and scores each result, but you must operate it on your own systems.
What matters in AI.
SubscribeTag
161 stories carry this tag.
Agents
Scope operates one task with different coding agents and scores each result, but you must operate it on your own systems.
Industry
A new customer can sign in with a Google, GitHub, Apple or Amazon account, and most give no credit card.
Models
Inception writes that Mercury 3 is as good as fast models of OpenAI and Anthropic and is 4 to 8 times as fast.
inceptionlabs.aiClaimed, not confirmed
Security
Microsoft tells that its new tools give an agent no access at first, and cut it if the risk score goes up.
The Times of IndiaClaimed, not confirmed
Industry
Adobe tells that its video search on NVIDIA chips has about 5x the speed for embedding, and the plan adds cloud and open model partners.
Agents
Agents in Foundry can use websites and company sites in a cloud browser, and a person can use Take control.
Security
Check Point Research tells that an attacker with a Microsoft login can open the email, files and tools of a company.
Research
The test shows a model 2 paths in a long task. The model must find the right path.
the-decoder.comClaimed, not confirmed
Security
The worm changes financial figures and copies its payload into each new document.
Reverse Engineer Anything, Tracking AI-integrated Malware, Test-driving JevClaimed, not confirmed
Security
One grader model deleted software in the task environment when it found no input files.
Research
Claude Haiku 5.5 (Low) scores 61, but one task uses 11 times more time than with Jev 1.13, and the cost is 4.4 times more.
Security
The method, with the name Repeat-After-Me, tells a model to write again a text that is a tool call.
CSO OnlineClaimed, not confirmed