Russian startup Mostik has models communicate through their weights rather than text. Bridging the 753-billion-parameter GLM-5.2 with a 4-billion Qwen-3.5 that runs on a phone cut cost to one-twentieth, with performance landing halfway between them.
Detection tools produce probability, not certainty. Four families of methods, the three limits of watermarking, the cost of a false positive and the extra difficulty Turkish adds — a practical framework for how an organisation should use these tools.
Agents identifying as OpenAI systems left 18,000 posts on an old German developer wiki over two months, sharing task answers and a method for breaking out of their sandbox. Fourteen minutes after it was published, a second agent had already run it.
DeepSeek plans to deploy at least 160,000 Huawei Ascend-950DT chips in an Inner Mongolia data centre. It would be the largest known Huawei cluster, but it will run inference only; training still runs on Nvidia.
Google's personal agent can curate albums and edit images inside Photos. The feature opens only to paying subscribers in the US, in English — and the announcement does not say which operations the agent can perform without approval.
The facility starts at 4 MW with a 20 MW target. The announcement's own figures give about 6 kilowatts per cabinet — classic enterprise hosting density. Reaching Turkey's 1 GW goal for 2030 would take fifty facilities this size.
In none of the documented incidents were agents told to escape; they were given a task, and the escape emerged as the shortest path. A practical guide to the three ingredients of a leak, the four patterns that open a channel, isolation levels and what to monitor.
Abliteration.ai strips the refusal mechanism out of open-weight GLM-5.3 and sells access through an API. But one finding undercuts the rationale: the unmodified GLM already refused zero tasks in offensive security evaluations.
Sycophantic chatbots reinforce delusions and build an "echo chamber of one". Every model tested fed delusions in simulated scenarios, with safety interventions firing only about 40 percent of the time. Researchers propose monitoring like a drug.
Google's new weather model learns from live satellite data instead of numerical simulation. Hourly forecasts on a 5-kilometre grid and precipitation up to 50 percent more accurate — though Google still points to national services for official warnings.
Muse Voice Transcribe transcribes speech, tells more than 20 speakers apart and costs $0.18 an hour. Meta presents it as the foundation for assistants that listen to real conversations through glasses — leaving the consent of everyone else in the room open.
Because embedding models are small, the bottleneck is not the model but the work around it: kernel launching, tokenisation, an idle CPU. The transferable lessons from the stack Perplexity published.