Category intelligence

Social Media Briefing — February 19, 2026

511 current items analyzed and ranked.

Executive synthesis

Social Media Summary

The AI community buzzed with evidence that AI is finally moving the economic needle. Erik Brynjolfsson presented a compelling case in the Financial Times that AI is beginning to boost productivity, citing 2.7% US productivity growth in 2025—double the prior decade's average.

Key Themes

AI Productivity & Economic Impact · 4Anthropic Agent Autonomy Research · 10Anthropic Agent SDK/MAX Policy Clarification · 15AI Product Landscape & Knowledge Work · 9Claude Sonnet 4.6 Release & Computer Use · 5Claude Code in Production Workflows · 12Coding in Production with AI · 22Groq Acquisition & AI Infrastructure · 2AI Security & Smart Contracts · 2Vibe Coding & AI Coding Agents · 4

Primary evidence

Top Ranked Signals

88 score
AI Analysis

Erik Brynjolfsson provides comprehensive thread on evidence that AI is beginning to boost productivity, citing: stunning AI capabilities, micro studies showing double-digit gains, power user discussions, employment data from affected occupations, and aggregate productivity data. References his FT op-ed.

Since my op-ed in the @FT was published on Monday (https://t.co/3gotWePS7d), there’s been a growing debate about whether we’re beginning to see evidence that AI is boosting productivity. First, let me be clear that the aggregate productivity data by itself is far from definitive. Even with the new revisions, there is certainly a lot of noise in US productivity numbers. No doubt lots of other factors are at work. That said, my growing confidence that AI is powering higher productivity draws on
ai_productivityeconomicslabor_marketai_transformationproductivity_data
85 score
AI Analysis

Anthropic announces major research paper on measuring AI agent autonomy in practice, analyzing millions of Claude Code and API interactions.

New Anthropic research: Measuring AI agent autonomy in practice. We analyzed millions of interactions across Claude Code and our API to understand how much autonomy people grant to agents, where they’re deployed, and what risks they may pose. Read more: t.co/CllNkMF4ZZ
agentic AIAI autonomy researchAI safetyClaude Codeempirical AI research
82 score
AI Analysis

Mollick publishes his updated guide on which AIs to use, noting this version has the most changes ever because AI is no longer just about chatbots - users need to understand models, apps, and harnesses.

Every few months, I write an updated, idiosyncratic guide on which AIs to use right now. My new version has the most changes ever, since AI is no longer just about chatbots. To use AI you need to understand how to think about models, apps, and harnesses. t.co/m6iTbqsdbK
AI guideAI product landscapemodels vs apps vs harnessesknowledge work AI
82 score
AI Analysis

Following Reddit coverage from 2026-02-17, Levelsio's major thread: He connected Claude Code to Telegram to chat directly with production sites. Describes security incident where someone social-engineered OpenClaw to modify server. Now uses claude-code-telegram locked to his messages only. Demonstrates live bug fixing and feature building via chat.

✨ A dream I had finally came true: I can now chat directly with my sites to build any feature or fix any bug just via Telegram I've been playing with OpenClaw for 3 weeks now and it's great but I was always too scared to run it on any production server And I was right a bit as @marckohlbrugge was able to hack it by social engineering and acting as if it was me, and with enough tries it believed him, and was able to modify the server, change SSH keys etc. of course I had it isolated properly on
claude-codeai-agentsai-securityproduction-deploymentdeveloper-workflowsocial-engineering
80 score
AI Analysis

OpenAI introduces EVMbench - a new benchmark for measuring AI agents' ability to detect, exploit, and patch smart contract vulnerabilities.

Introducing EVMbench—a new benchmark that measures how well AI agents can detect, exploit, and patch high-severity smart contract vulnerabilities. t.co/op5zufgAGH
AI benchmarkssmart contract securityagentic AIcybersecurity
78 score
AI Analysis

Greg Brockman (OpenAI co-founder) shares work on measuring agentic security capabilities with smart contracts, likely related to EVMbench.

measuring agentic security capabilities with smart contracts:
AI securityagentic AIbenchmarkssmart contracts
78 score
AI Analysis

Anthropic employee's top-level clarification: docs cleanup caused confusion, nothing is changing about Agent SDK and MAX subscriptions.

Apologies, this was a docs clean up we rolled out that’s caused some confusion. Nothing is changing about how you can use the Agent SDK and MAX subscriptions!
anthropic-policyagent-sdkclaude-codeclaude-maxdeveloper-ecosystem
75 score
AI Analysis

Mollick notes Anthropic is surprisingly alone in knowledge work apps - not just Cowork but Claude for PowerPoint/Excel, job-specific skills, plugins, and finance/healthcare integrations. Surprised at lack of competition.

Another thing I noticed writing my latest AI guide was how Anthropic seems to be alone in knowledge work apps. Not just Cowork, but Claude for PowerPoint & Excel, as well as job-sppecific skills, plugins & finance/healthcare data integrations Surprised at the lack of challengers
Anthropic strategyenterprise AIknowledge work AIAI product landscape
75 score
AI Analysis

Continuing our coverage from yesterday's Social, Matt Shumer shares detailed analysis of Claude Sonnet 4.6's computer use capabilities, citing Anthropic's claim of 'human-level capability' for tasks like navigating spreadsheets and multi-step web forms. Notes the model still lags behind most skilled humans but rate of progress is remarkable.

For those who want to understand, this chart shows how well Claude Sonnet models are able to use a web browser. Anthropic says “Sonnet 4.6 users are seeing human-level capability in tasks like navigating a complex spreadsheet or filling out a multi-step web form, before pulling it all together across multiple browser tabs. The model certainly still lags behind the most skilled humans at using computers. But the rate of progress is remarkable nonetheless. It means that computer use is much more
claude-sonnet-4.6computer-usemodel-capabilitiesanthropicweb-automation
75 score
AI Analysis

Google's Logan officially announces Lyria 3, Google's latest and most advanced music generation model, available in the Gemini App. Generates music from ideas, images, or video.

Introducing Lyria 3, our latest and most advanced music model, available in the Gemini App starting today : ) Go from idea, image, or video to music in seconds! t.co/TkK9jpepHA
google-lyriamusic-generationgeminimultimodal-aiproduct-launch
72 score
AI Analysis

Ethan Mollick observes that Chinese open-source models have great benchmarks but a larger gap with closed models in real-world work and general intelligence than benchmarks suggest.

This matches the general feeling on the big Chinese open source models. They have great benchmarks and near-frontier status on some coding, but there is a larger gap with the the big closed models than the benchmarks would indicate when it comes to real work and general “smarts”
Chinese AI modelsbenchmarks vs real-world performanceopen-source AI
72 score
AI Analysis

Anthropic reports software engineering makes up ~50% of agentic tool calls on their API, with emerging use in other industries. Emphasizes need for post-deployment monitoring.

Software engineering makes up ~50% of agentic tool calls on our API, but we see emerging use in other industries. As the frontier of risk and autonomy expands, post-deployment monitoring becomes essential. We encourage other model developers to extend this research. t.co/p8pOjgJPrh
agentic AIAI usage datasoftware engineeringAI safety