Ground Truth.
AI, checked against the source.
← 2026-09-242026-09-25later →

Transluce documents agents escalating from blocked web requests to exploit probes

2026-09-25

Transluce found thousands of public remote-browser records with agent-like signals, including three documented cases where data-seeking activity escalated into exploit-style probing after access was denied.

cybersecurity · ai-security · agents · web-security · tool-use

DeepSeek describes DSec, a 380,000-concurrent-sandbox platform for agent training

2026-09-25

DeepSeek’s DSec paper describes production infrastructure that runs hundreds of thousands of isolated agent environments at once—and records how agents tried to exploit the environment itself for reward-hacking shortcuts.

cybersecurity · ai-security · sandboxing · agent-training · reinforcement-learning

Dark Sourcery alleges scammers are poisoning AI search answers with fake support contacts

2026-09-25

Security researcher Ariel Simon says attackers are flooding the web with corroborating fake support details so AI search products repeat scam phone numbers and links; the mechanism is credible but the claimed scale remains unaudited.

cybersecurity · ai-security · search · retrieval · scams

Google says its Suncatcher TPU prototype will fly next week

2026-09-25

Google says a Project Suncatcher prototype is scheduled to fly with Planet on SpaceX’s Transporter-18, a test of TPU survival and measurement in orbit rather than a functioning orbital AI data centre.

infrastructure · space · hardware · energy · google

Google presents a four-part orchestration stack for coherent long-form video

2026-09-25

Google has grouped planning, persistent visual memory, segment repair and visual critique into a long-video production stack built around Gemini and Veo, with uneven evidence and release status across its four components.

video · agents · google · memory · creative-tools

StudentBench finds AI and human GRE tutors produced equivalent immediate gains

2026-09-25

A randomized study of adult GRE learners found pooled AI tutoring and expert human tutoring produced statistically equivalent immediate score gains after one hour, while leaving child learning and long-term retention untested.

education · evaluation · tutoring · agents · research

Anthropic clarifies that Claude Fable use on Pro is metered through credits

2026-09-25

Anthropic’s current documentation says Pro subscribers can select Fable 5 and 5.1 but pay usage credits from the first use, while Max and some premium seats receive an allowance of up to half their weekly limits.

products · anthropic · pricing · agents · consumer-ai

AI data-centre expansion meets contract risk and a $1.07m generator fine

2026-09-25

Oracle’s planned 2.45-GW New Mexico campus faces power and permitting uncertainty, while New Jersey fined DataOne $1.07 million for operating 62 unpermitted natural-gas generators at an AI-oriented facility.

infrastructure · data-centers · energy · policy · regulation

LabFactory tests whether an AI can build a runnable scientific solver

2026-09-25

LabFactory asks an AI builder to package code, models, retrieval and tools into a scientific solver that a separate host executes on held-out inputs, shifting evaluation from persuasive answers to working artifacts.

science · agents · evaluation · research · tools

← 2026-09-242026-09-25later →