On September 29, 2026, OpenAI will preview GPT-6 Cyber at its annual DevDay conference in San Francisco. It will be the company’s fourth cybersecurity-focused model released in twelve months. The cadence alone is the story: GPT-5.4 Cyber arrived in April, GPT-5.5 Cyber in June, GPT-5.6 Cyber in August, and now GPT-6 Cyber in September. No other frontier lab has shipped domain-specific models at anything approaching this frequency. But the model itself is not the most significant announcement. The infrastructure around it is. — Read More
Recent Updates
The plunging price of thought
AI has gotten cheaper more quickly than any other transformative technology in history. The cost of achieving a given level of AI performance has fallen about 47% per quarter since 2023, or 13× per year. That price drop is four times faster than DNA sequencing, six times faster than compute, 18 times faster than lithium batteries, and (in the century up to 1973) 54 times faster than electricity.
… It is now widely understood that the AI boom is a macroeconomic force powerful enough to raise prices for the inputs it demands: chips, power, even the labor of electricians. Less well recognized is a paradoxical flip side: the price of the output from all those data centers is falling extraordinarily rapidly. — Read More
DeepSeek Elastic Compute (DSec): A Sandbox Infrastructure for Effective Agentic Training at Scale
Large-scale agentic training and evaluation with large language models (LLMs) rely on isolated, stateful execution environments in which models inspect repositories, invoke tools, execute commands, and interact with task-specific services. These workloads create sandboxes in large bursts, span heterogeneous functionality and isolation requirements, retain state across long interactions, and draw from large image corpora with limited reuse. Supporting them therefore requires an elastic execution platform rather than a single sandbox runtime. This report presents DeepSeek Elastic Compute (DSec), a production sandbox platform that exposes FnCall, container, microVM, and full-VM sandbox backends through a unified SDK. DSec coordinates placement and lifecycle management across the cluster, composes environments from independently versioned layers, combines memory sharing, reclamation, and CPU scheduling for high-density execution, and loads image data on demand from Fire-Flyer File System (3FS), a cluster-wide distributed filesystem. DSec is co-designed with the reinforcement learning (RL) framework, decouples stateful rollout execution from preemptible GPU training, coordinates sandbox lifecycle with training to preserve rollout state while reclaiming idle resources, and mitigates agent misbehavior such as reward hacking. A single production-scale unit of DSec spans around 160 nodes, serving about 3 million sandboxes per day; in production, it supports over 380,000 concurrent sandboxes and sustains over 5,000 sandbox creations per second. Our evaluation and deployment experience show that these mechanisms reduce environment setup and image-distribution overhead, improve memory efficiency, and preserve latency-sensitive performance under high-density overcommit. — Read More
Self-Jailbreaking: Language Models Can Reason Themselves Out of Safety Alignment After Benign Reasoning Training
We discover a novel and surprising phenomenon of unintentional misalignment in reasoning language models (RLMs), which we call self-jailbreaking. Specifically, after benign reasoning training on math or code domains, RLMs will use multiple strategies to circumvent their own safety guardrails. One strategy is to introduce benign assumptions about users and scenarios to justify fulfilling harmful requests. For instance, an RLM reasons that harmful requests like outline a strategy for stealing customers' credit card information from a retail store'' could be associated with the benign intent of a security professional trying to test defense,” despite no such benign context being provided as input. We observe that many open-weight RLMs, including DeepSeek-R1-distilled, s1.1, Phi-4-mini-reasoning, and Nemotron, suffer from self-jailbreaking despite being aware of the harmfulness of the requests. We also provide a mechanistic understanding of self-jailbreaking: RLMs are more compliant after benign reasoning training, and after self-jailbreaking, models appear to perceive malicious requests as less harmful in the CoT, thus enabling compliance with them. To mitigate self-jailbreaking, we find that including minimal safety reasoning data during training is sufficient to ensure RLMs remain safety-aligned. Our work provides the first systematic analysis of self-jailbreaking behavior and offers a practical path forward for maintaining safety in increasingly capable RLMs. — Read Morehttps://arxiv.org/pdf/2510.20956
Anthropic’s Misuse Report, Condensed to 117 Findings
Eight months of disrupted AI abuse, read case by case and boiled down to a 20-page document with every claim linked to its source.
Anthropic published its September 2026 threat intelligence report on September 10. It covers the misuse they disrupted between December 2025 and August 2026, and it is long. We read the whole thing and condensed it into 117 findings across eight categories, each one a single sentence, and each one linking back to the exact passage it came from. — Read More
MCP Is Dead: Why This Protocol Breaks in Production(And How to Fix It)
30 CVEs in 60 days, a maintenance tax nobody warned you about, and what engineers are quietly switching to.
… MCP isn’t going away. The downloads are real. The Linux Foundation governance is serious. Multi-vendor adoption means the protocol has institutional staying power.
But the MCP of early tutorials — install a community server, plug it in, done — that version is dead. — Read More
Learning from Real-World Experience
How users interact with our products in the real world is a valuable source for model training. The data is abundant, reflects the actual distribution of user tasks, and captures user corrections and tool failures that synthetic environments may miss.
A common way to learn from real-world data is rejection sampling fine-tuning: judge each session’s outcome, keep the successful ones, and train the model to imitate them. But a successful outcome does not mean every step was correct, so imitating the whole trajectory risks reinforcing bad intermediate behaviors in addition to good ones. Discarding unsuccessful sessions also loses critical evidence of where the model falls short.
We combine rejection sampling fine-tuning with hint-guided self-distillation to learn from both successful and unsuccessful sessions. A hint is a short corrective instruction grounded in information the model already had when it made the mistake. Useful steps from successful sessions remain imitation targets, while grounded hints turn avoidable mistakes into correction targets. In live use, the later trained checkpoint reduced tool-call failures by 21.2% relative to an earlier trained checkpoint. — Read More
DeepSeek-V4.1-Flash: Frontier Agents on a Smaller Memory Budget
A lot of recent progress in language models has focused on better reasoning, larger reinforcement-learning runs, and increasingly sophisticated agent training. One of the many remaining bottlenecks is the context. Coding agents, research agents, and tool-using systems repeatedly process hundreds of thousands of tokens while generating comparatively little output. At that point, prefill compute and the KV cache become the most important infrastructure problems.
DeepSeek-V4.1-Flash is essentially an attempt to redesign the Transformer around that workload. Its 552B-parameter MoE backbone uses a new Causal Encoder-Decoder (CED) architecture, activating only about 8B parameters per token during prefill and 16B during decoding. Compressed Sparse Attention 2 (CSA2) shares KV representations and sparse-attention decisions across layers; SWA Bounded Replay avoids persistently storing sliding-window KV states; and FP4 KV quantization pushes the global cache down to only 890 bytes per token — about one quarter of DeepSeek-V4-Flash’s. The model still supports a 1M-token context, native vision, and strong agentic capabilities. It is trained on 45T multimodal tokens, with much of the improvement coming from scaling automatically generated agent tasks, environments, and rollouts. — Read More
Frontier Overhangs
There has been, over the last week, what I think is a healthy debate about the philosophy and psychology that undergirds the views of meaningful segments of the AI community, particularly those obsessed with doomsday scenarios. It is, in the end, difficult to reason with a philosophy that grants equivalent moral weight to not just all beings — human or not — who exist today, but who may ever exist in the future; this tilts the scales in such an absurd fashion towards safetyism that innovation is impossible and freedom is intolerable.
Worse, it taps into the psychology of religion, where dissent is not brooked and questioning the premise is heresy. I guess that makes me a heretic then: I reject the premise in favor of doubt in our ability to foresee the future, combined with faith in humanity figuring things out along the way. — Read More
dlab Open Source Week: Frontier AI on Your Own Hardware
In one of my classes I asked the question I was afraid to ask but I just needed the answer to: “Who is afraid of not getting a job after graduating?” About eighty percent of the 150 people in the room raised their hands. … The other story arrives by email. PhD students who cannot wait to graduate, because they want to join a frontier lab and they have concluded that research in academia is meaningless. They are counting the years until they can leave.
I believe both stories are wrong, and wrong for the same reason. They assume the future of research belongs to whoever has the most GPUs. I think the opposite is true. Academia is probably about to have a renaissance, and the most exciting work of the next decade will happen in university labs — not in spite of their limited resources, but because of them. — Read More
