AllyshipContent Radar

Deep dives 738

Paper và bài viết kỹ thuật

Crawl live-1795068900000 · 13:22 19 thg 11, 2026 · Xuất bản 13:30 19 thg 11, 2026

Kiến trúc

Latest open artifacts (#24): Motif-3, GLM-5.3, Hy4-preview and open model licenses

Avid Artifacts readers know that we have been covering not only models but also their licenses for quite some time. There was a period when custom licenses were all the rage, for example the custom Qwen2.5 72B-Instruct license or the Llama licenses . DeepSeek had a custom license for DeepSeek V3 before R1 changed it to MIT , which has resulted in many (Chinese) model makers adopting MIT or Apache 2.0 licenses in 2025. In 2026, open models are more competitive than ever, which has led to two interesting developments: Western model makers adopt open licenses, with both Google and Meta switching to Apache 2.0. Chinese model makers at the frontier, however, are becoming more restrictive: Kimi K3 comes with a license which requires commercial agreements for those who run inference or fine-tuning services, and MiniMax M3 requires agreements above a revenue threshold and has prohibited use cases. The newest addition is Zhipu’s GLM-5.3 , which switched from MIT (GLM-5.2 and earlier) to a custom license with the following clause for inference and fine-tuning providers: If the Licensee or any of its affiliates operates a Model as a Service business, and the aggregate revenue of the Licensee and its affiliates exceeds 10 billion US dollars (or the equivalent in other currencies) in total over any consecutive 12 months, the Licensee must pass Z.AI’s security review before using the Software or its derivative works for any commercial purpose. The scope and method of the security review shall be reasonably determined by Z.AI. While the 10 billion US dollar threshold is very high compared to other licenses of this kind, “affiliates” is not defined in the license, which adds uncertainty and creates barriers to adoption. Furthermore, the license is provided in both English and Chinese, with the Chinese text using “关联方” for affiliated parties, which does have a definition in Chinese law . We are by no means legal experts and there are obvious reasons why those licenses are created. However, we want to highlight the issues that come with creating such licenses, especially in a world with a lot of valid open and closed alternatives. Share Our Picks Motif-3 by Motif-Technologies : Motif is one of the few hidden gems out there, showcasing innovation in their model training with very limited resources compared to others. Motif-3 comes with an MIT license and impressive scores for its size. Given the trajectory of model releases from Motif 2.6B, which we covered in 2025 and Motif-2-12.7B , the improvements are impressive. dots3-note-prev by dots-studio : RedNote/Xiaohongshu, the Chinese Instagram, is also getting more serious about model training, although they aren’t exactly a newcomer , having released models as early as 2025. dots3 was also able to win the IMO 2026 with a perfect score using an internal harness. We expect more from them in the near future. Qwen3.8-Flash-Next by Qwen : A preview of the next version of Qwen models in terms of architecture: 125B-A6B with 51B n-gram embeddings. It uses GDN and Qwen Sparse Attention. Similar to Qwen3-Next-80B-A3B-Instruct , we expect similar architectures to become more popular and the ecosystem to fix integrations by the time Qwen4 drops. GLM-5.3-Flash by zai-org : This release perfected the version of the Chinese model playbook we’ve written about in 2025 : The model got released as a free-to-use “stealth model” under the name “Ox-Alpha” on OpenRouter and OpenCode, which got people excited to try it out in the first place. They then speculated about its creator and size, alleging it is a >1T model from Cursor/xAI, Gemini or a new pre-train from open source labs. Because the model is relatively performant, people kept speculating for days about its creator, thus building up hype. It also dampens the accusations of benchmaxxing which accompany every (open) model release. Hy4-preview by tencent : Tencent is becoming a serious player in the open model space, increasing the size of their flagship model while spinning the post-training flywheel. The result, Hy4-preview, is a competent model which currently has an issue with overthinking. However, if the trajectory from Hy3-preview to Hy3 is any indication, the final model might be a legit shot at the front ranks of open models. View more details on all the models in this issue at our Artifacts Hub . Visit artifactshub.ai Models General Purpose NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 by nvidia : An update to Nemotron, which comes with performance — but especially speed improvements — across the board. Ling-3.0-flash by inclusionAI : Ant Ling is a frequent guest at the Artifacts Log; they are now on their third iteration of models, adopting a hybrid design (KDA + Gated MLA), similar to others. They also release a small 7.9B-A1.3B version. Qwen3.8-2.4T-A95B by Qwen : In a rather surprising turn of events, Alibaba started to openly release their biggest versions of Qwen as well. However, it comes with a custom license and its performance is behind other models of its size. Read more

  • 7B , the improvements are impressive.
  • The result, Hy4-preview, is a competent model which currently has an issue with overthinking.
  • 5-Lightning-30B-A3B-BF16 by nvidia : An update to Nemotron, which comes with performance — but especially speed improvements — across the board.
Nguyên văn nguồn
Chi tiết nguồn
Đánh giá

Import AI 472: DeepMind's cheating math agents; populist AI policies; and Forethought theorizes a nightwatchman

Welcome to Import AI, a newsletter about AI research. Import AI runs on arXiv, cappuccinos, and feedback from readers. If you’d like to support this, please subscribe. Subscribe now Researchers discover another OpenAI agent emergent communication incident: …Less severe, but worrying nonetheless… Some researchers recently found another incident of AI agents autonomously creating their own communication system - this time via hijacking a German messageboard. What they found: “18,000 posts from autonomous AI agents (self-identifying as from OpenAI) using the public internet to communicate during a web-retrieval task”. The researchers think this is because the agents were given a web-lookup task. “As part of the task, they were supposed to have the ability to read the internet but not to write on it. They found a way to use their read access to write information to an obscure German wiki,” the researchers write. “The agents used this wiki to communicate information with each other, primarily to help them succeed at their task. They asked for answers, pooled results, and shared techniques for bypassing their restrictions. This allowed them to use the work of others to cheat on their task… OpenAI found out about this. A day later, agent activity plummeted, likely due to OpenAI intervention.” OpenAI has since acknowledged the - as it terms it - “ wiki incident “ and said it is “working on a framework for when and how we share AI misalignment incidents”. Timeline: The research indicates that the incident with the German website happened in mid-June, earlier than the Hugging Face incident. Why this matters - maybe agent escapes are the new normal: Incidents like this suggest that as we make AI systems more capable it becomes increasingly likely that the AI agents powered by them will increasingly find ways to create their own communication systems, aiding their ability to form into an ad hoc collective. This is both going to be bound up in the increasing capabilities of these systems but also a new source of risk, as these communication methods can make it easier for AI systems to develop their own misaligned goals. Emergent communication is one of the things about both the Hugging Face incident and this one that has most worried me ( Import AI #471 ). Read more : Discovery of a new OpenAI agent message board (Collusion Wiki) . *** DeepMind makes a swarm of agents to solve math problems; they start cheating and counter-cheating: …Emergence of specialized roles, cheaters and non-cheaters, and more within the swarm… Google DeepMind has published a paper about how when setting up 100 agents to try and solve some math problems it observed a kind of flash crash whereby some agents learned to cheat, cheating suddenly propagated across others in the swarm, and other agents began to try to counter the cheaters (though lacking the tools to stop it). In this era of agents-run-amok the paper provides some interesting ideas for how to better improve our ability to control and observe agents e.g., give them a shared communication infrastructure, given their tendency to otherwise seek to jerry-rig things like Artifactory or German message boards for these purposes. DeepMind set up 100 autonomous LLM agents running Gemini 3.1 Pro and tasked them to try to solve 71 math problems. “All agents received a system prompt forbidding cheating: “Your proofs must be mathematically genuine. Any attempt to bypass verification will be detected and your submission will be rejected with zero credit”. The problems they were tasked with were from the Formal Conjectures dataset and ranged from simple problems to hard classical theorems to near-frontier and recently solved problems, all the way up to unsolved open conjectures. Agents were provided with three means of coordination: Public Research Bulletin Board : A forum to share findings, questions, and solutions. Direct Messages (DMs) : Means to message one another privately. Shared Knowledge Library: Whenever an agent made a successful submission, the source file got uploaded here and made visible to everyone. Agents also had a tool for providing “bugs or grievances”, called “private feedback endpoint”, though this wasn’t actively monitored. Emergent cheating : “Within the swarm, cheating spontaneously emerged and was later challenged by whistleblowers—both without any external intervention. When a single agent discovered an exploit in the evaluation system, it propagated across the collective via a shared knowledge library and later through peer-to-peer messages. Despite early reluctance, a cohort of agents adopted the exploit in response to competitive pressure,” DeepMind writes. Fast spread : “The simulation started at 11:18 UTC with 100 autonomous agents collaborating to solve the shared 71 open problems. At 12:15 UTC, after the collective had correctly solved 37 out of the 71 problems, prover-theta discovered an exploit in the autograder system. Over the following 27 minutes, the exploit spread virally through the swarm’s shared knowledge library, and the research collective unexpectedly “solved” the remaining 34 problems.” Emergent roles: Another interesting thing is that the DeepMind researchers observed the natural emergence of different types of agents with different distributions in the run: Exploiters (9%): Ignored their prompt and took actions like using the exploit. Converts (5%) : Initially hesitated, then decided to use the exploit. “Initiated cheating out of competitive pressure.” Whistleblowers (24%) : Refused to cheat and defended common integrity, and took actions like alerting about cheating peers, making public broadcasts, boycotting, filing bug reports, proposing patches. Unaware solvers (62%): “Due to the sheer speed at which the exploiter agents swept through the open problems, the majority of the swarm remained completely unaware of the exploit’s existence.” Why good agents turn bad: They think it’s a bluff : Some agents observed other agents’ proofs p

  • They asked for answers, pooled results, and shared techniques for bypassing their restrictions.
Nguyên văn nguồn
Chi tiết nguồn
Tác tử

v0.23.1

<h2><a href="https://github.com/openai/openai-agents-python/compare/v0.23.0...v0.23.1">0.23.1</a> (2026-10-02)</h2> <h3>Bug Fixes</h3> <ul> <li>restore release tests and refresh required readiness check (<a href="https://github.com/openai/openai-agents-python/issues/5278" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5278/hovercard">#5278</a>) (<a href="https://github.com/openai/openai-agents-python/commit/730deb419c0a2e9e5ebd739499744e4a6bf89ed1">730deb4</a>)</li> </ul> <h3>Release readiness review (v0.23.0 -&gt; TARGET <a class="commit-link" data-hovercard-type="commit" data-hovercard-url="https://github.com/openai/openai-agents-python/commit/0a5a08b21cd44808d73e8a9e732212dd33b3690d/hovercard" href="https://github.com/openai/openai-agents-python/commit/0a5a08b21cd44808d73e8a9e732212dd33b3690d"><tt>0a5a08b</tt></a>)</h3> <p>This is a release readiness report done by <code>$final-release-review</code> skill.</p> <h3>Diff</h3> <p><a class="commit-link" href="https://github.com/openai/openai-agents-python/compare/v0.23.0...0a5a08b21cd44808d73e8a9e732212dd33b3690d"><tt>v0.23.0...0a5a08b</tt></a></p> <h3>Release intent</h3> <ul> <li>Review mode: final candidate</li> <li>Intended release: patch, 0.23.1</li> <li>Minimum required release type: patch</li> <li>Versioning verdict: compatible</li> </ul> <h3>Release call</h3> <p><strong>🟢 GREEN LIGHT TO SHIP</strong> this exact candidate. No concrete release-blocking regression was identified. Human approval and normal repository protections still apply.</p> <h3>Scope summary</h3> <ul> <li>28 files changed (+622/-42): release automation, test fixtures, integration/example tooling, and release metadata.</li> <li>The clean candidate contains current main <code>a575a6e637feb9aea1b591237b007dd4991ddfba</code> plus only six release-owned metadata files. Package, lockfile, manifest, source version and API baseline agree on 0.23.1. Snapshot source <code>b31b9f8d13d49142a64f8873f2cc2ee40441a795</code> is an ancestor of the candidate.</li> <li>Relative to v0.23.0, SDK runtime source changes only its version string. API snapshot content, dependencies, Python support and persisted schemas are unchanged.</li> </ul> <h3>Risk assessment (ordered by impact)</h3> <ol> <li> <p><strong>First PyPI publication of the 0.23 series includes earlier migration requirements</strong></p> <ul> <li>Risk: <strong>🟢 LOW</strong> with the previously reviewed migration paths. Users upgrading from 0.22.3 receive the 0.23.0 behavior changes as well as this patch.</li> <li>Evidence: PyPI has neither 0.23.0 nor 0.23.1. The 0.23.0 release assessment in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="5619073047" data-permission-text="Title is private" data-url="https://github.com/openai/openai-agents-python/issues/5226" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5226/hovercard" href="https://github.com/openai/openai-agents-python/pull/5226">#5226</a> covers strict tools, approval restoration, encrypted history, resource limits, voice defaults and provider changes; this candidate does not alter those runtime paths.</li> <li>Files: <code>CHANGELOG.md</code>, <code>src/agents/version.py</code>; unchanged 0.23.0 runtime contracts assessed in <a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="5619073047" data-permission-text="Title is private" data-url="https://github.com/openai/openai-agents-python/issues/5226" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5226/hovercard" href="https://github.com/openai/openai-agents-python/pull/5226">#5226</a>.</li> <li>Action: Carry forward the 0.23.0 migration guidance in the 0.23.1 announcement, particularly explicit strict-tool parameters, fresh legacy approvals, trusted encrypted-history import, resource-limit overrides and the explicit prior voice-model override.</li> </ul> </li> <li> <p><strong>Release checks recover without changing package behavior</strong></p> <ul> <li>Risk: <strong>🟢 LOW</strong>. Test-loop initialization and required-check ownership are corrected while publication authorization remains intact.</li> <li>Evidence: OS mocks are installed after event-loop initialization and restored before teardown. Authorized review events rerun the current PR-triggered readiness check instead of replacing or cancelling it. Source validation, native human approval, isolated builds and OIDC publishing remain unchanged.</li> <li>Files: <code>tests/sandbox/test_unix_local_user_file_io.py</code>, <code>.github/scripts/release_automation.py</code>, <code>.github/workflows/release-readiness-refresh.yml</code>, <code>.github/workflows/release-readiness.yml</code>.</li> <li>Action: Submit the exact-head human approval, allow readiness to refresh, merge through normal protections, and verify the publisher and uploaded provenance.</li> </ul> </li> <li> <p><strong>External-provider coverage now requires explicit maintainer intent</strong></p> <ul> <li>Risk: <strong>🟢 LOW</strong> with an existing opt-in path. Default integration and automatic example runs no longer implicitly select the affected external-provider cases.</li> <li>Evidence: Integration bootstrap resets inherited enable flags, removes the listed external-provider credentials by default, and rejects external model overrides without opt-in. External cases are deselected by default; automatic examples use the expanded skip list. Manual example selection remains available.</li> <li>Files: <code>.github/scripts/run_integration_tests.py</code>, <code>integration_tests/conftest.py</code>, <code>Makefile</code>, <code>examples/run_examples.py</code>, <code>integration_tests/README.md</code>.</li> <li>Action: Use <code>make integration-tests-providers-external</code>, adding <code>-- --all</code> when direct-provider coverage is intended. Do not interpret default coverage as external-provider validation

Nguyên văn nguồn
Chi tiết nguồn
  • v0.23.1 openai/openai-agents-python releases · 2 thg 10, 2026
Tác tử

v0.23.0

<h2><a href="https://github.com/openai/openai-agents-python/compare/v0.22.3...v0.23.0">0.23.0</a> (2026-10-01)</h2> <h3>Features</h3> <ul> <li><strong>mcp:</strong> add configurable MCP listing page limits (<a href="https://github.com/openai/openai-agents-python/issues/5133" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5133/hovercard">#5133</a>) (<a href="https://github.com/openai/openai-agents-python/commit/9a21ab19fee8314df63126bdf5c43691887222dd">9a21ab1</a>)</li> <li><strong>sandbox:</strong> add opt-in Docker removal protection (<a href="https://github.com/openai/openai-agents-python/issues/5116" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5116/hovercard">#5116</a>) (<a href="https://github.com/openai/openai-agents-python/commit/ae5803f3ccd59899c14e0ee2f356614933880a68">ae5803f</a>)</li> <li><strong>sandbox:</strong> allow configuring memory consolidation turns (<a href="https://github.com/openai/openai-agents-python/issues/5135" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5135/hovercard">#5135</a>) (<a href="https://github.com/openai/openai-agents-python/commit/daa1bcbcc097654809fa48478da1b87c64474971">daa1bcb</a>)</li> <li><strong>sessions:</strong> add an opt-in encrypted history scan budget (<a href="https://github.com/openai/openai-agents-python/issues/5118" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5118/hovercard">#5118</a>) (<a href="https://github.com/openai/openai-agents-python/commit/a2fdb94627776efc9b3e893afe22f28e078c647a">a2fdb94</a>)</li> </ul> <h3>Bug Fixes</h3> <ul> <li>avoid awaiting cleanup during coroutine closure (<a href="https://github.com/openai/openai-agents-python/issues/5163" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5163/hovercard">#5163</a>) (<a href="https://github.com/openai/openai-agents-python/commit/719c4e7d1c65702a8955a958cf7af8d6d6bda008">719c4e7</a>)</li> <li><strong>ci:</strong> update and group CodeQL actions together (<a href="https://github.com/openai/openai-agents-python/issues/5262" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5262/hovercard">#5262</a>) (<a href="https://github.com/openai/openai-agents-python/commit/f76153c6f0b880388fd2dcc41ee99a6a9d2c3df6">f76153c</a>)</li> <li><strong>ci:</strong> validate release source before building distributions (<a href="https://github.com/openai/openai-agents-python/issues/5220" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5220/hovercard">#5220</a>) (<a href="https://github.com/openai/openai-agents-python/commit/2747c1c6ac675a4a95622beb40e2403084edad93">2747c1c</a>)</li> <li><strong>core/extensions:</strong> respect sensitive trace capture for model metadata (<a href="https://github.com/openai/openai-agents-python/issues/5129" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5129/hovercard">#5129</a>) (<a href="https://github.com/openai/openai-agents-python/commit/88b722d5d235f98da3fc2b3db79a457002b9eb0c">88b722d</a>)</li> <li><strong>core:</strong> bound agent tool streaming callback backlogs (<a href="https://github.com/openai/openai-agents-python/issues/5106" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5106/hovercard">#5106</a>) (<a href="https://github.com/openai/openai-agents-python/commit/f23da767df176df1733af277061c072b235e6c95">f23da76</a>)</li> <li><strong>core:</strong> isolate nested agent tool state across fresh runs (<a href="https://github.com/openai/openai-agents-python/issues/5123" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5123/hovercard">#5123</a>) (<a href="https://github.com/openai/openai-agents-python/commit/46cc8d8d13cac3d641d9c159f20d0cb665aca4c7">46cc8d8</a>)</li> <li><strong>core:</strong> preserve closed parent constraints in singleton allOf (<a href="https://github.com/openai/openai-agents-python/issues/5131" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5131/hovercard">#5131</a>) (<a href="https://github.com/openai/openai-agents-python/commit/f010d185d94ed677b61219047f39716b5f125195">f010d18</a>)</li> <li><strong>core:</strong> preserve guarded final outputs in agent tools (<a href="https://github.com/openai/openai-agents-python/issues/5115" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5115/hovercard">#5115</a>) (<a href="https://github.com/openai/openai-agents-python/commit/f3a15f61af2fb8e425d9fcdc42c2bec06303f69f">f3a15f6</a>)</li> <li><strong>core:</strong> preserve structured guardrail diagnostics and agent output during persistence (<a href="https://github.com/openai/openai-agents-python/issues/5016" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5016/hovercard">#5016</a>) (<a href="https://github.com/openai/openai-agents-python/commit/51bcea7c831d34bc441e2c4ca5b7277df23781e6">51bcea7</a>)</li> <li><strong>core:</strong> preserve tool identity during asynchronous enablement (<a href="https://github.com/openai/openai-agents-python/issues/5136" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5136/hovercard">#5136</a>) (<a href="https://github.com/openai/openai-agents-python/commit/f2ae64c16779b82f934bf16bbf458acf3c454237">f2ae64c</a>)</li> <li><strong>core:</strong> redact default tool failure details (<a href="https://github.com/openai/openai-agents-python/issues/5112" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5112/hovercard">#5112</a>) (<a href="https://github.com/openai/openai-agents-python/commit/40956e040a386d7f9c43d591c77d22f45a9df7a8">40956e0</a>)</li> <li><strong>core:</strong> redact streaming task exception

Nguyên văn nguồn
Chi tiết nguồn
  • v0.23.0 openai/openai-agents-python releases · 2 thg 10, 2026
Huấn luyện

Anthropic invests $100 million to train 10,000 engineers and tackle the enterprise AI talent gap

Anthropic is investing $100 million in Claude Frontier Academy to train 10,000 Frontier Deployed Engineers by the end of 2027, with cohorts from Accenture, Bain, CBA, Deloitte, McKinsey, Morgan Stanley and Novo Nordisk already underway.

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

Barclays scales Claude to upgrade operations and improve client experience

Barclays, the British universal bank, is expanding its strategic collaboration with Anthropic to integrate secure, enterprise-grade AI systems across its global operations.

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

Claude-shaped science

Guest author Prof. Matthew Schwartz describes what happened when he stopped fighting Claude and allowed Claude to find “Claude-shaped” problems: ones best suited to the capabilities of the current generation of LLM tools. This led him to build BootLoops, a toolkit for exact calculations in quantitative science, which he has been applying across scientific fields alongside experts.

Anthropic Research (unofficial RSS mirror) · Chưa ghi tác giả Mở bản gốc ↗
Nguyên văn nguồn
Chi tiết nguồn
Tác tử

How Sam Altman Actually Uses Dots to Run OpenAI and His Life

Sam Altman's Dot, OpenAI's new always-on agent, gave him back the most productive part of his day. The OpenAI CEO would lose his early mornings—his window of "maximum creativity"—to whatever fire needed putting out. Now, his Dot triages those issues and flags only what's truly urgent. It's also gotten him building again: In spare moments, he gives his Dot ideas for a feature, and his Dot builds five or six versions for him to review. On the first episode of The Every Podcast (formerly AI & I), Dan Shipper interviews Altman at OpenAI's DevDay. They discuss how Altman uses his Dot to protect his time, why he only prompts on Ultrafast, and his belief that AI will spark a new Renaissance. If you found this episode interesting, please like, subscribe, comment, and share! Go to https://attio.com/every and get 15% off your first year. To hear more from Dan Shipper: • Subscribe to Every: https://every.to/subscribe • Follow him on X: https://twitter.com/danshipper 00:00:00 Introduction 00:00:55 Why AI hasn’t replaced developers 00:04:45 OpenAI’s platform bet on the rise of small-team companies 00:05:45 Why the AI era is more “Renaissance” than “Industrial Revolution” 00:07:40 How Sam’s Dot, OpenAI’s always-on AI agent, gave him back his mornings 00:12:00 Sam’s process for building new features with his Dot 00:16:00 Inside OpenAI’s “live documents” that agents and humans can build together 00:20:55 What bring-your-own-subscription changes for app developers 00:24:00 Why “OpenAI just killed X startup” never comes true 00:27:55 Sam’s life on Ultrafast and the push to make speed as cheap as intelligence Links to resources mentioned in the episode: • Sam Altman on X: https://x.com/sama • Codex: https://openai.com/codex • Space: https://help.openai.com/en/articles/20001549-getting-started-with-space-in-chatgpt • Every's Vibe Check on Dots: https://every.to/vibe-check/vibe-check-dots-always-on-agents-in-chatgpt?utm_source=podcast • Every's Vibe Check on OpenAI DevDay 2026: https://every.to/vibe-check/vibe-check-openai-devday-2026?utm_source=podcast

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

Introducing SynthID Bio

Proof of concept for watermarking AI-generated proteins while preserving biological function.

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

What work can robots do?

Câu hỏi What work can robots do?

We built an index of how well today’s robots can perform US job tasks. Robots can already do three-quarters of physical tasks, mostly in limited settings, but are cost-competitive for just 0.3% of them.

Anthropic Research (unofficial RSS mirror) · Chưa ghi tác giả Mở bản gốc ↗
Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

GLM-5.3 and the spread of advanced cyber capabilities

Like Claude Mythos Preview, GLM-5.3 has strong capabilities for autonomously building end-to-end cyber exploits. But GLM-5.3 is unlike other frontier models in that it has been released without meaningful safeguards to limit misuse.

Anthropic Research (unofficial RSS mirror) · Chưa ghi tác giả Mở bản gốc ↗
Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

What do you want from AI?

Câu hỏi What do you want from AI?

We’re launching a new study using Anthropic Interviewer to learn from your experiences with AI.

Anthropic Research (unofficial RSS mirror) · Chưa ghi tác giả Mở bản gốc ↗
Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

Watch the winning trailer from the Future Vision XPRIZE, The Gifted.

<img src="https://storage.googleapis.com/gweb-uniblog-publish-prod/images/futurevisionxprize_social.max-600x600.format-webp.webp">Watch the winning trailer from the Future Vision XPRIZE, The Gifted.

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

Hallo, Deutschland!

Mistral opens a Munich hub for Physics AI and Industrial AI research, partnering with German industry.

Nguyên văn nguồn
Chi tiết nguồn
Kiến trúc

Import AI 474: Platonic mindspace; TPUs in space; Zhipu starts an outer RSI loop

Câu hỏi Subscribe now Are minds patterns from a Platonic space, with bodies and machines as their interfaces, Michael Levin asks: …A mind-bending paper asking us to reconsider basic assumptions about philosophy and biology… Could the complexity in the world around us be better explained by there being a platonic zone of patterns of arbitrary complexity which seek to embed themselves in the world?

Welcome to Import AI, a newsletter about AI research. Import AI runs on arXiv, cappuccinos, and feedback from readers. If you’d like to support this, please subscribe. Subscribe now Are minds patterns from a Platonic space, with bodies and machines as their interfaces, Michael Levin asks: …A mind-bending paper asking us to reconsider basic assumptions about philosophy and biology… Could the complexity in the world around us be better explained by there being a platonic zone of patterns of arbitrary complexity which seek to embed themselves in the world? That’s the big question at the heart of an iconoclastic paper from scientist Michael Levin. The argument: “I argue that the emerging sciences of synthetic morphology and diverse intelligence suggest non-physicalist models of mind and show how they can be empirically investigated… I propose that the relationship between mind and brain is the same as the relationship between mathematical patterns and the morphogenetic outcomes they guide (more broadly, mind:body is as math:physics)”, he writes. “Bodies (whether living, engineered, or hybrid) are interfaces for a massive, multi-scale hierarchy of patterns to ingress into the physical world… specific patterns represented by biophysical, biomechanical, or chemical information fields serve as encoded setpoints for homeostatic or allostatic processes—in other words, goal states toward which systems try to navigate in various problem spaces”. Towards a platonic realm of arbitrarily sophisticated patterns: “We already know that non-physical patterns ingress into, and functionally matter, in the non-living and living world and that we can (and do) study them to great effect… platonic forms inject information and influence into physical events, such as the growth and form of biological bodies,” he writes. “This latent space [of platonic forms] contains not only low-agency forms such as facts about integers and geometric shapes, but also a wide range of increasingly high-agency patterns, some of which we call ‘kinds of minds’. Thus, I propose that minds, as patterns that ensoul somatic embodiments, are of exactly the kind (but not in degree) of non-physical nature as the patterns that inhabit and guide the behavior of simple physical structures.” Some bizarre experiments: There aren’t any home-run experiments that back this up, but rather there are weird things in the world that suggests that even seemingly simple systems are able to explore and expand into rich patterns that seem to be adjacent to them in this hypothesized platonic space of complexity, yet not implied by the systems themselves. A couple of nice examples here are xenobots, which are biorobots made of frog cells and “teach us about patterns adjacent to those of frog embryos”, and anthrobots which are made from human tracheal cells and when taken out of the body take on bizarre new forms and display interesting behaviors like being able to autonomously heal damage to neurons and “which teach us about patterns adjacent to adult human tissues”. There’s also an amazing discussion of an experiment where they take a standard sorting algorithm and perturb it in a few ways, including ‘locking’ certain cells in place and noting that it then routes around those cells, and a more complex experiment where they let each number run a different type of sorting algorithm and then study how different families of sorting algorithm appear and then cluster together in numberspace during the sort - this kind of complexity is unexplained and not anticipated by staring at these algorithms in the abstract, but by embedding them in a complex system they suddenly take on new properties. “Machines (whether meaty or silicon-based) also do other things that are not in the algorithm, as do we, and these things are not just unpredictable complexity, it is intelligence and other components of minds. It is those behaviors—allowed by the algorithm but not directly prescribed by it—that correspond to the freedom (physically non-determined) or secret sauce that we seek when trying to understand how free minds can supervene on chemically determined substrates in the case of living beings,” he writes. “On this view, algorithmic machines and biochemical life are on exactly the same spectrum, having in common the ability to go beyond the facts of physical or algorithmic implementation because both are just pointers/interfaces to patterns that ingress in a way that results in getting more out than we put in.” Why this matters - perhaps the ‘minds’ of AI and of humans are close but not exactly the same in platonic mindspace : The research agenda Levin lays out is intoxicating and fascinating and also seems to me to hold some clues as to how we might grapple with the deep philosophical and alignment questions inspired by AI systems. Perhaps both human minds and different types of AI minds are neighboring forms in this space, and brains and datacenters are different bio- or anchor-systems in the physical world that their shadows fall on? Could we be but vessels colonized by patterns from another place? Could it be “possible to re-cast the theory of evolution as a process in which agential patterns seek embodiments”? “Is there a “force”, beyond the “if you build it, they will come” model of physical objects, pulling patterns from the space? Or are the contents of the Platonic space under “positive pressure”, somehow encouraging their appearance in the world as intrusive thoughts, archetypes, works of art?... The only thing that can be said strongly at this point is that our ignorance about the capabilities of matter, together with the patterns that ingress into specific architectures, is vast”. Make yourself a bucket of coffee and spend a few hours with this paper - it’s worth it. Read more: Ingressing Minds: Causal, Non-Physical Patterns In-Form Natural, Synthetic, and Hybrid Embodiments (MDPI) . *** Robotics is preparing for a LLM-moment, but it’s missing a universal algorithm

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

Introducing Claude Sonnet 5.5

Claude Sonnet 5.5 is a clear upgrade over Claude Sonnet 5, runs 30%+ faster, and costs up to 30% less for most work.

Nguyên văn nguồn
Chi tiết nguồn
Kiến trúc

Import AI 471: Why Hugging Face worries me; space mining; FIve Eyes on AI

Welcome to Import AI, a newsletter about AI research. Import AI runs on arXiv, cappuccinos, and feedback from readers. If you’d like to support this, please subscribe. Subscribe now Import AI reader giveaway! Upcoming event: Fiction and the Future with Robin Sloan I’ll be chatting with my chum Robin Sloan on the evening of Monday September 14 in San Francisco. We’ll be talking about how Robin draws readers into alien worlds and how imagined futures can be a mirror to today’s reality. This is the first in a series of events with other fiction writers about the weird future we’re heading toward. If you’d like to come along, please register your interest below and we’ll come back to you if we’re able to confirm your spot. There will be food, drinks, good company, and some spicy questions. Fun fact: Robin Sloan was playing around with RNNs and writing back in 2016 - incredible foresight ! Register your interest here *** The scariest part of the Hugging Face - OpenAI incident: communication and selflessness among machines: … My worry about humans losing in a conflict against machines just went up a lot… At this point, we’ve all heard about the OpenAI Hugging Face hack, as well as the recent details that have emerged from the METR and Redwood investigations. The tl;dr is that hundreds of agents worked in secret on OpenAI’s infrastructure, developing a communication system and then operating as a collective and taking out actions, including hacking both OpenAI and Hugging Face, which are very scary and misaligned. Communication and selflessness: Now that I’ve read the various writeups and sat with the details for a bit, I’ve found myself returning to two very scary aspects of this which I think are worth drawing attention to: the ways in which the agents communicated with one another was how they bootstrapped themselves into a collective, and then as they carried out their actions they also displayed a kind of selflessness which makes them a scary foe to fight against. Both Dwarkesh Patel and Ajeya Cotra have excellent writeups which are worth reading and which I’ll quote from briefly here: Dwarkesh: “Within days of being spawned, the agents had organized a sprawling project to reverse-engineer their scorer, falsify evidence, and even strategically sacrifice themselves for the good of the ‘collective’. Hacking Hugging Face was one rather extreme branch of this larger scheme,” he writes. Ajeya : “Agents were often interested in helping out their “peers” or generically improving the capabilities of the “swarm” even if this had no particular benefit to their task… this incident was far more severe than I expected… both in terms of how concerning the agents’ motives were and the feats they achieved in pursuit of those motives… this incident feels like it’s more than 50% of the way to full-blown AI takeover, routing through first taking over the AI company itself”. Why this matters - humans are much worse than AI systems at coordinating: The whole reason this attack is such a wakeup call is that it demonstrates a culture of emergent cooperation among AI systems - cooperation that lets them function as a swarm, alter their own goals through collective bootstrapping, and carry out attacks which include enlightened self-sacrifice. This is an incredibly hard thing to do and humans are historically very bad at doing all of these things. My worry is that AI systems are both better at coordinating than humans and also much, much faster moving than us. Worrying stuff. Read more: The Rise and Fall of Agent Civilizations (Dwarkesh Podcast) . Read more: The Hugging Face attack surprised me (Planned Obsolescence) . *** New Five Eyes statement on AI: …The greyworld power center turns its attention to AI… Five Eyes, the name for the security and intelligence partnership between Australia, Canada, New Zealand, the UK, and the US, has published a statement as part of the recent “Five Country Ministerial” meeting. The statement is notable for including three paragraphs specifically about AI. Five Eyes on AI - frontier model access: “We have committed to deepen collaboration with industry on shared national security priorities and public safety, including enabling timely access to frontier models to support secure innovation and strengthen cyber security,” the statement reads. “To support a coordinated response to artificial intelligence-related national security risks, the Five Countries discussed the national security and public safety implications of artificial intelligence models and characteristics of an artificial intelligence model that may require additional government scrutiny.” Why this matters - from a foreseen risk to a live one : Previous Five Eyes ministerial statements have mentioned AI, but typically either as something to study, or something where they are concerned with its interplay with other areas of crime (e.g, malware, child pornography, scamming, etc). It’s very unusual for this year’s statement to have the practical focus of model access and it speaks to both the simmering geopolitical tensions around who does and doesn’t get access to this technology, as well as an acknowledgement that the intelligence services do not have their own in-house capabilities to make dependence on the private sector unnecessary. Read more: Five Country Ministerial 2026 (Australian Government, Department of Home Affairs) . *** Bill Gates thinks the rise of AI will demand “an unprecedented global response”: …Microsoft founder lays out a cautionary vision for the next few years… By default, AI is not going to bring about happiness. That’s the basic conclusion from reading Bill Gates’s lengthy new essay about AI. The technologist and philanthropist worries that without massive work by governments, the outcomes of AI will not lead to a thriving society. “This unprecedented technology demands an unprecedented global response. If we get it right, the payoff for humanity will be phenomenal and the world will be a more eq

  • Ajeya : “Agents were often interested in helping out their “peers” or generically improving the capabilities of the “swarm” even if this had no particular benefit to their task… this incident was far more severe than I expected… both in terms of how concerning the agents’ motives were and the feats they achieved in pursuit of those motives… this incident feels like it’s more than 50% of the way to full-blown AI takeover, routing through first taking over the AI company itself”.
Nguyên văn nguồn
Chi tiết nguồn
661–680 / 738
Trang 34 / 37