TLDR AI 2026-02-16

OpenClaw (formerly ClawdBot and MoltBot) is an open-source agent assistant that operates with the same permissions as the user who installs it - and can act on their behalf.

With over 120,000 GitHub stars and 20,000 forks, OpenClaw is so popular because of its efficacy - it's risky for the same reason.

>> Get your bearings on OpenClaw security by watching Zenity's on-demand webinar

>> Want to get certified in AI security? With theFoundations of AI Security series, you can learn how to deploy secure AI across the enterprise directly from experts at Zenity and AWS. Attend all three virtual sessions to earn earn the Foundations of AI Security Professional Certificate!

Register here

OpenClaw's creator will join OpenAI to develop accessible agents and ensure OpenClaw remains open and independent within a new foundation. OpenAI's support will provide access to cutting-edge models and resources, aligning with the creator's vision of making these tools widely usable. OpenClaw's community will benefit from a supportive structure that encourages innovation and data ownership.

OpenAI introduced an optional Lockdown Mode for higher-risk workflows and added “Elevated Risk” labels for capabilities in ChatGPT, ChatGPT Atlas, and Codex that can increase exposure to prompt-injection attacks.

xAI's Grok Build plans to introduce Parallel Agents and Arena Mode, enhancing its capabilities from a coding assistant to a full IDE. Parallel Agents allows up to eight coding agents to work simultaneously, while Arena Mode facilitates competition among agents to determine the best output. The update also includes a new UI resembling a browser-based IDE, dictation support, and nonfunctional GitHub integration, with Grok 4.20 expected soon.

Both CEOs of OpenAI and Anthropic claim that human-level AI is just around the corner. However, Transformers have their fundamental limitations. While they are very powerful and have taught us a lot about what general intelligence is, overcoming these problems will require research that could take decades. The advances that made modern GPTs possible were discovered gradually over several decades, and each discovery came with new and more nuanced issues.

Anthropic and OpenAI have unveiled new "fast modes" for LLM inference, offering differing performance enhancements. OpenAI's model speeds exceed 1,000 tokens per second using Cerebras chips, but rely on a less capable model, while Anthropic's fast mode supports real models with up to 2.5x speed through low-batch-size inference. Despite OpenAI's technical achievement, the utility of faster but less capable inference remains debatable, focusing more on OpenAI's exploration of Cerebras' potential.

Text has to pass through a tokenizer before GPT-5.x is able to understand any input. A tokenizer is a compression layer that converts raw text into a sequence of integer IDs. Every design decision baked into the tokenizer has downstream consequences for cost, accuracy, multilingual performance, and hallucination rates. This article looks at OpenAI's open-source tokenizer library, tiktoken, and explains how it works.

OpenClaw's recent security issues highlight a larger problem with agent ecosystems as new attack surfaces, where malicious skills can exploit agent + tools + marketplace combinations. Current safeguards like prompts are inadequate. Agents need sandboxing, scoped credentials, restricted tools, and proper logging to manage risks.

Human judgment is the backbone of AI quality. Yet most orgs scaling AI lack clear decision frameworks, consistent interpretation policies, and reliable oversight mechanisms. Welo Data builds AI data quality systems that govern human and machine judgment. Turns out that judgment works when it's designed, not assumed. Talk to an expert

Personaplex is a real-time speech-to-speech conversational model that jointly performs streaming speech understanding and speech generation. It operates on continuous audio and can predict both text tokens and audio tokens auto-regressively to produce its spoken responses. Personaplex's design allows the model to update its internal state based on users' ongoing speech while still producing fluent output audio, so it can support highly interactive conversations. The model is ready for commercial use.

Minimax M2.5 is fast, cheap, and great at coding. While the team behind the model has released a technical article on how the model works, it is quite dense. This post aims to make the content of that article more accessible for readers. The core problem the team is solving is how to make RL work at massive scale for being good at Agentic AI. Training LLMs to be good at agentic tasks imposes many challenges: huge volumes of training data need to be processed quickly, training needs to be stable, and agents need to be good at a diverse range of tasks.

AI agents can tackle increasingly complex tasks, but to achieve more ambitious goals, they need to be able to meaningfully decompose problems into manageable sub-components and safely delegate their completion across to other AI agents and humans. This study proposes an adaptive framework for intelligent AI delegation. The proposed framework is applicable to both human and AI delegators and delegatees in complex delegation networks.

Composition-RL reused “too-easy” pass-rate-1 prompts by automatically combining multiple problems into new verifiable questions for RLVR training. Across 4B–30B models, the approach improved reasoning over training on the original dataset.

You have to know where you're going before you take that first step. The first thing to do is always to define your goal. Reverse engineer requests and combine that with context to create discrete, boolean, testable Ideal State Criteria. This criterion can carry through to become the verification criteria as well.

Anthropic's partnership with Palantir means that the Defense Department and federal law enforcement have access to Claude. This means it was possible that the technology was used in matters related to the US military's operation to capture former Venezuelan President Nicolás Maduro. Anthropic's usage guidelines prohibit Claude from being used to facilitate violence, develop weapons, or conduct surveillance. A company spokesperson said that Anthropic has not discussed the use of Claude for specific operations, but the company is committed to using frontier AI to support the US in national security.

Microsoft plans to reduce its reliance on OpenAI by developing its own AI models, spearheaded by AI chief Mustafa Suleyman.

An example of how an agent can be used to edit parts of software like a wiki.

Manus AI launched a 24/7 Agent via Telegram but faced immediate suspension without explanation.

openrouter/free is a router that selects free models at random from the models available on OpenRouter.

The exponential growth of AI is due to a series of bottlenecks being smashed, and Groq's LPU is the next step.