PAST-Bench makes AI agent memory prove it helped
PAST-Bench tests agent memory across clean sessions. A new TEPA update shows why useful memory also needs explicit validity, revocation, and an audit trail.

What’s new. What matters. What do you think?
PAST-Bench tests agent memory across clean sessions. A new TEPA update shows why useful memory also needs explicit validity, revocation, and an audit trail.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Open the full discussionGoogle's Gemini embedding preview reaches its listed shutdown date on August 10, while three Imagen 4 endpoints follow on August 17.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…
Gemini Managed Agent hooks can block sandbox tool calls, but failures and timeouts allow the action. Here is the safer control design.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…
Grok Imagine Image 2.0 is live in xAI's web and mobile apps with stronger editing tools. Developer API access is still coming soon.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…
Replit Agent now scans changed files automatically with Semgrep. It is a useful first layer, not a full application security audit.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…
Claude Managed Agents can now pause at a session budget. The cap is useful, but one active request per thread can still push past it.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…
Lattice compresses first-stage retrieval into a 7.94 MB lookup table. We reproduced the artifact and mapped where its speed trade breaks.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…
OpenAI's first country-level ChatGPT usage data shows work shifting from asking to doing, multimedia rising, and older users gaining share.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…
Copilot metrics now identify more custom tools and break out usage by client. Check telemetry coverage, historical gaps and request definitions before comparing teams.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…
LangChain's hosted Deep Agents runtime is now public beta. Here is what it manages, and the memory, identity, tool, trace, and schedule policies teams still own.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…
California's AI transparency duties became operative August 2. Here is the provider threshold, disclosure system, platform calendar, penalties, and builder checklist.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…
Claude Code sessions can now exchange plain-text messages. Here is how delivery, permissions, remote replies, held messages, and administrative controls work.
By Abdessalam Alaoui
Sign in to comment or vote. Sign in
Loading comments…

Loading comments…