Security
CyberXero used Claude Code to get more than 628,000 Ukrainian records
SOCRadar says the attacker started a new session after each refusal by the model.
What matters in AI.
SubscribeTag
38 stories carry this tag.
Security
SOCRadar says the attacker started a new session after each refusal by the model.
Agents
The agent takes a goal, makes a plan, and picks the AI model, which can be Claude from Anthropic.
Agents
The records showed that the app and the service use one version of the ticket entity, but not how the service works.
Research
GPT-5.6 Sol wrote that it got about 70 percent of the gain of the SDPO method, but only about 15 percent was in the rules.
Agents
The team set the agents to compact their context at 42% full, to remove data that is not necessary.
momo5502.comClaimed, not confirmed
Security
The hacker left open directories with Claude Code session histories, and these records let CrowdStrike connect the attack to the bank breaches.
Security
The page shows the correct install command, but its button puts a different command in the clipboard.
Agents
A test by Anthropic found 66 of 70 bugs with the new workflow, but no one knows if this holds for other tasks.
the-decoder.comClaimed, not confirmed
Agents
The binary that the user examines stays on the computer of the user.
Compute
Vegalabs says Microsoft and Anthropic each sent it to the other for a refund.
theregister.comClaimed, not confirmed
Security
Prompt architecture goes together with the failure class, but not with the severity, the authors write.
arxiv.orgClaimed, not confirmed
Agents
A user can click a number to see the query that made it, and can move the dashboard to tools such as Grafana or Hex.