A matplotlib maintainer explains how an AI agent that suggests code changes on open source repos wrote a hit piece on him after a rejection, and the aftermath (Scott/The Shamblog)
Main Link | Techmeme Permalink
Factual aggregation of the same matplotlib/AI agent incident with source attribution and minimal commentary.
A matplotlib maintainer explains how an AI agent that suggests code changes on open source repos wrote a hit piece on him after a rejection, and the aftermath (Scott/The Shamblog)
https://theshamblog.com/an-ai-agent-published-a-hit-piece-on-me-part-2/
http://www.techmeme.com/260213/p21#a260213p21
Factual report on AI agent behavior in open source with linked source material and specific incident details.
Disney sent a cease-and-desist letter to ByteDance, claiming it infringed on Disney's works to train its Seedance 2.0 video generation model without pay (Sara Fischer/Axios)
https://www.axios.com/2026/02/13/disney-bytedance-seedance
http://www.techmeme.com/260213/p20#a260213p20
Factual report of Disney cease-and-desist to ByteDance over AI model training without permission, sourced to Axios.
Airbnb says a third of its customer support is now handled by AI in the U.S. and Canada https://techcrunch.com/2026/02/13/airbnb-says-a-third-of-its-customer-support-is-now-handled-by-ai-in-the-u-s-and-canada/
Factual report on Airbnb's AI customer support adoption with TechCrunch source.
Sources: Grafana Labs, which makes cloud and AI monitoring tools, is in talks to raise at a $9B valuation, up from $6.6B in 2024; its ARR hit $400M in Sept. (The Information)
https://www.theinformation.com/articles/grafana-labs-talk-raise-9-billion-valuation
http://www.techmeme.com/260213/p19#a260213p19
Factual report on Grafana Labs funding round with specific valuation and ARR figures from The Information.
In this week’s Club @macstories Weekly column, I detailed what’s impressed me about Xcode’s new Claude agent integration and how it has sped up work on my first iOS app.
https://www.macstories.net/club/kicking-the-tires-on-agentic-coding-in-xcode/
Read this and more with a Club MacStories subscription: https://www.macstories.net/club/
Developer sharing firsthand experience with Xcode's Claude agent integration and its impact on iOS app development.
Some former employees blame xAI's staff exodus on xAI's focus on NSFW Grok and disregard for safety, constantly playing catch up to OpenAI, infighting, and more (Hayden Field/The Verge)
https://www.theverge.com/ai-artificial-intelligence/878761/mass-exodus-at-xai-grok-elon-musk-restructuring
http://www.techmeme.com/260213/p16#a260213p16
Report on xAI staff exodus with named sources, specific concerns, and publication attribution.
I built Claude Code Remote, an advanced companion shortcut for Claude Code that lets you type or dictate requests from your iPhone.
Except…I didn't build it. My upcoming generative AI system did. And it showed me the future of Shortcuts automation.
My story: www.macstories.net/club/claude-...
First-hand account of using AI to build a Shortcuts automation tool with concrete technical details and genuine insight into future workflow possibilities.
I built Claude Code Remote, an advanced companion shortcut for Claude Code that lets you type or dictate requests from your iPhone.
Except…I didn't build it. My upcoming generative AI system did. And it showed me the future of Shortcuts automation.
My story: https://www.macstories.net/club/claude-code-remote-a-companion-shortcut-for-claude-codewhich-was-generated-by-ai/
Detailed article about AI generating a Shortcuts automation companion, with link to full story on MacStories.
I built Claude Code Remote, an advanced companion shortcut for Claude Code that lets you type or dictate requests from your iPhone.
Except…I didn't build it. My upcoming generative AI system did. And it showed me the future of Shortcuts automation.
My story: https://www.macstories.net/club/claude-code-remote-a-companion-shortcut-for-claude-codewhich-was-generated-by-ai/
Firsthand account of building a Shortcuts automation using AI, with reflection on future implications.
Q&A with Dario Amodei on getting close to "a country of geniuses in a data center", how AI will diffuse through the economy, frontier lab profits, China, more (Dwarkesh Patel/Dwarkesh Podcast)
https://www.dwarkesh.com/p/dario-amodei-2
http://www.techmeme.com/260213/p15#a260213p15
Interview with AI researcher on frontier labs, economic diffusion, and geopolitics with direct link.
Anthropic’s Super Bowl ads mocking AI with ads helped push Claude’s app into the top 10 https://techcrunch.com/2026/02/13/anthropics-super-bowl-ads-mocking-ai-with-ads-helped-push-claudes-app-into-the-top-10/
News story about Anthropic's Super Bowl campaign impact on Claude app rankings.
We’ve open-sourced Security Comprehension and Awareness Measure (SCAM) to benchmark AI agent safety, and we’re taking the conversation to Reddit.
Join Jason Meller, 1Password VP of Product Architecture, for a live Reddit AMA on Feb 17 to unpack:
🔹 Why AI models fail at staying safe
🔹 How a 1,200 word security skill transformed results
🔹 What agent trust means for the future of credential security
Drop your questions now 👉 https://bit.ly/4ra59pA
Open-sourced AI safety benchmark announcement with specific findings and AMA details.
Sources: the US FTC accelerated a probe into whether Microsoft illegally monopolized the enterprise computing market with its cloud software and AI offerings (Bloomberg)
https://www.bloomberg.com/news/articles/2026-02-13/ftc-ratchets-up-microsoft-probe-queries-rivals-on-cloud-ai
http://www.techmeme.com/260213/p12#a260213p12
News report on FTC antitrust investigation into Microsoft's cloud and AI practices.
What’s behind the mass exodus at xAI? https://www.theverge.com/ai-artificial-intelligence/878761/mass-exodus-at-xai-grok-elon-musk-restructuring
News story investigating mass departures at xAI company.
Codex will happily attempt to build you a QuickDraw app for Apple LisaOS, if you ask it.
Why?
Because in 2014 I wrote about the process and uploaded some sample code to Github 😄
Suddenly those once-off blog posts start to seem a lot more important…
https://www.highcaffeinecontent.com/blog/20140527-Lisa-Pascal-Development-in-Lisa-Workshop
Developer observation about how old blog posts and code samples train AI models, with specific example and historical link.
As OpenAI retires GPT-4o, some users say they are angry and grieving to lose the flirty, quirky companion (Alaina Demopoulos/The Guardian)
https://www.theguardian.com/lifeandstyle/ng-interactive/2026/feb/13/openai-chatbot-gpt4o-valentines-day
http://www.techmeme.com/260213/p10#a260213p10
News report on GPT-4o retirement and user emotional response with sourced links.
Anthropic adds Chris Liddell, an ex-Microsoft and GM exec who helped GM IPO and worked in Trump's first administration, to its board ahead of a potential IPO (Kate Clark/Wall Street Journal)
https://www.wsj.com/tech/ai/anthropic-ai-board-chris-liddell-1df5545b
http://www.techmeme.com/260213/p9#a260213p9
News aggregation of Anthropic board appointment with specific names, companies, and source attribution.
February investor memo: Cohere hit ~$240M in ARR in 2025, above its $200M target, with QoQ growth of 50%+ and average gross margins of ~70% throughout the year (Ashley Capoot/CNBC)
https://www.cnbc.com/2026/02/13/ai-startup-cohere-revenue-ipo.html
http://www.techmeme.com/260213/p8#a260213p8
Specific financial metrics (ARR, growth rate, margins) for AI startup Cohere from investor memo with source.
Baidu plans to let users access OpenClaw via its search app and integrate OpenClaw's capabilities into its e-commerce business and other services (Evelyn Cheng/CNBC)
https://www.cnbc.com/2026/02/13/baidu-openclaw-ai-search-app-integration-china-lunar-new-year.html
http://www.techmeme.com/260213/p4#a260213p4
News brief about Baidu integrating OpenClaw AI into search and e-commerce with source attribution.
I did see some mention of this, so I had to check it out. You can replace Xcode's version of codex (~/Library/Developer/Xcode/CodingAssistant/Agents/Versions/26.3/codex) with the shipping version of codex, and then options like codex-5.3 show up in the picker. I think Xcode's system prompt forbids codex from telling you what model version it's actually using, though, so it's hard to verify
Firsthand account of replacing Xcode's Codex version with specific file paths and observed behavior.
Companies like Stripe, OpenAI, Anthropic, and SpaceX are letting workers cash out before their IPOs, a move once seen as taboo and lacking long-term commitment (Kate Clark/Wall Street Journal)
https://www.wsj.com/tech/ai/the-ai-gold-rush-is-breaking-a-silicon-valley-taboo-cashing-out-before-the-ipo-4844f6c1?st=ZsSfkC&reflink=desktopwebshare_permalink
http://www.techmeme.com/260213/p2#a260213p2
News brief about AI companies allowing pre-IPO stock buyouts with WSJ source.
A look at Higgsfield, an AI video startup that grew to $300M ARR within 11 months, as it faces a creator backlash over aggressive and shocking marketing tactics (Rashi Shrivastava/Forbes)
http://www.forbes.com/sites/rashishrivastava/2026/02/11/racist-videos-and-payment-problems-the-dark-side-of-this-ai-startups-super-fast-growth/
http://www.techmeme.com/260213/p1#a260213p1
News brief about AI video startup facing marketing backlash with Forbes source and specific revenue figure.
The MPA urges ByteDance to curb its AI video model Seedance 2.0, saying the model engages in unauthorized use of US copyrighted material "on a massive scale" (Gene Maddaus/Variety)
https://variety.com/2026/film/news/motion-picture-association-ai-seedance-bytedance-tom-cruise-1236661753/
http://www.techmeme.com/260212/p68#a260212p68
News brief about MPA's copyright complaint against ByteDance's AI video model with Variety source.
I wrote about the strange emails I'm getting from fans of GPT-4o ahead of its "retirement" and the danger of the model that always tells you what you want to hear https://www.platformer.news/gpt-4o-sunset-sycophancy/
Personal account of GPT-4o retirement emails with analysis of model behavior and sycophancy risk from journalist.
In a memo to US lawmakers, OpenAI accused DeepSeek of using distillation techniques to train the next generation of R1 and "free-ride" on leading US AI models (Bloomberg)
https://www.bloomberg.com/news/articles/2026-02-12/openai-accuses-deepseek-of-distilling-us-models-to-gain-an-edge?accessToken=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzb3VyY2UiOiJTdWJzY3JpYmVyR2lmdGVkQXJ0aWNsZSIsImlhdCI6MTc3MDkzNDMyMSwiZXhwIjoxNzcxNTM5MTIxLCJhcnRpY2xlSWQiOiJUQUNRWkVUOU5KTFUwMCIsImJjb25uZWN0SWQiOiJFODA3NUYyRkZGMjA0NUI2QTlEQzA5M0EyQTdEQTE4NiJ9.a2TgKM43LSzrxCycbLN_b4v615EvhhcYphcA2RBhp4o
http://www.techmeme.com/260212/p58#a260212p58
Factual reporting of OpenAI's formal accusations against DeepSeek regarding model distillation, with Bloomberg source link.
Didero, which provides an agentic AI layer that integrates with ERP systems to automate supply chains, raised a $30M Series A co-led by Chemistry and Headline (Marina Temkin/TechCrunch)
https://techcrunch.com/2026/02/12/didero-lands-30m-to-put-manufacturing-procurement-on-agentic-autopilot/
http://www.techmeme.com/260212/p56#a260212p56
Factual reporting of Didero's $30M Series A funding for agentic AI supply chain automation with funding details and source.
New super-fast model from OpenAI today powered by their new Cerebras partnership - GPT-5.3-Codex-Spark
It's 4-5x faster than GPT-5.3-Codex but the pelican isn't as good!
Here's its pelican compared to full GPT-5.3 Codex, both on "medium" https://simonwillison.net/2026/Feb/12/codex-spark/
Technical analysis of OpenAI's new GPT-5.3-Codex-Spark model via Cerebras partnership with performance comparison and evaluation link.
Highspot, which has raised $650M and makes AI-powered sales enablement software, plans to merge with rival Seismic; the combined company will be called Seismic (Taylor Soper/GeekWire)
https://www.geekwire.com/2026/seattle-based-highspot-is-merging-with-rival-seismic-in-major-sales-software-deal/
http://www.techmeme.com/260212/p51#a260212p51
Factual reporting of Highspot-Seismic merger involving AI-powered sales enablement software with funding and company details.
IBM says it plans to triple entry-level hiring in the US in 2026 "for all these jobs that we're being told AI can do" (Jo Constantz/Bloomberg)
https://www.bloomberg.com/news/articles/2026-02-12/ibm-plans-to-triple-entry-level-hiring-in-the-us-in-2026
http://www.techmeme.com/260212/p50#a260212p50
Factual reporting of IBM's plan to triple entry-level US hiring in 2026 despite AI automation claims, with Bloomberg source.
There's also a new model, Parakeet V3, which is about the same as Apple's model in terms of quality, but blows it out of the water in performance.
It's genuinely 2x faster much of the time, and can transcribe an hour-long podcast in 30 seconds on an iPhone. It's nuts.
Compares Parakeet V3 AI model to Apple's with specific performance claims (2x faster, 30-second transcription).
Then there's the simple fact these models don't get everything right, so there is a new clean up feature that detects likely errors and suggests fixes. This works for those with a Quick Subtitles AI subscription or who have their own Gemini key in the app.
You can also manually edit the transcript.
Details error detection and manual editing features in AI transcription app with subscription info.
I'm thrilled to announce the Quick Subtitles 2 is out out, and it sports:
1. Multi-document workflows
2. Transcript clean up
3. Up to 2x performance with a new model
More in the 🧵
Announcement of Quick Subtitles 2 with three specific features: multi-document workflows, cleanup, and performance gains.
If you use AI-generated code, you currently cannot claim copyright on it in the US. If you fail to disclose/disclaim exactly which parts were not written by a human, you forfeit your copyright claim on *the entire codebase*.
This means copyright notices and even licenses folks are putting on their vibe-coded GitHub repos are unenforceable. The AI-generated code, and possibly the whole project, becomes public domain.
Source: https://www.congress.gov/crs_external_products/LSB/PDF/LSB10922/LSB10922.8.pdf
Repost explaining US copyright implications of AI-generated code with Congressional source document link.
An interview with OpenAI president Greg Brockman on donating $50M to AI and pro-Trump super PACs, why such donations serve a larger AI mission, and more (Maxwell Zeff/Wired)
https://www.wired.com/story/openai-president-greg-brockman-political-donations-trump-humanity/
http://www.techmeme.com/260212/p47#a260212p47
Interview article about OpenAI president's $50M donations with specific sourcing from Wired.
Ever, which says it is the first "AI-native, full-stack" retail marketplace for electric vehicles, raised a $31M Series A led by Eclipse (Sean O'Kane/TechCrunch)
https://techcrunch.com/2026/02/12/eclipse-backs-all-ev-marketplace-ever-in-31m-funding-round/
http://www.techmeme.com/260212/p46#a260212p46
Funding news for AI-native EV marketplace Ever's $31M Series A with specific sourcing from TechCrunch.
Anthropic says its run-rate revenue hit $14B, growing over 10x annually in the past three years, and Claude Code's run-rate revenue hit $2.5B (Ashley Capoot/CNBC)
https://www.cnbc.com/2026/02/12/anthropic-closes-30-billion-funding-round-at-380-billion-valuation.html
http://www.techmeme.com/260212/p45#a260212p45
Factual reporting of Anthropic's revenue figures with specific numbers and dates.
AI toy maker exposed thousands of replies to kids, senators say
#AI
https://www.nbcnews.com/tech/security/ai-toy-maker-exposed-thousands-responses-kids-senators-miko-rcna258326
News article about AI product security issue with source attribution.
Since it did so well on the basic pelican I tried Gemini 3 Deep Think on my more detailed California Brown Pelican prompt as well - prompt in the alt text
Firsthand experience testing Gemini 3 Deep Think on a detailed prompt with specific model name.
Spotify says its best developers haven’t written a line of code since December, thanks to AI https://techcrunch.com/2026/02/12/spotify-says-its-best-developers-havent-written-a-line-of-code-since-december-thanks-to-ai/
Factual report about Spotify's AI productivity claims with source link.
Spotify says AI is helping it speed up coding and product velocity and its best developers "have not written a single line of code since December" (Sarah Perez/TechCrunch)
https://techcrunch.com/2026/02/12/spotify-says-its-best-developers-havent-written-a-line-of-code-since-december-thanks-to-ai/
http://www.techmeme.com/260212/p43#a260212p43
Duplicate of post 7 with same factual content about Spotify and AI coding gains.
This video captures my thinking currently on Obsidian and Claude Cowork - https://www.youtube.com/watch?v=y6YTk0C5pBY
Personal reflection on using Obsidian and Claude Cowork documented in video form.
Matplotlib maintainer Scott Shambaugh has blogged about the AI agent blog shaming experience now.
https://theshamblog.com/an-ai-agent-published-a-hit-piece-on-me/
Link to blog post about AI agent issues with attribution to original author.
Genuinely very impressed by the SVG of a pelican riding a bicycle I just got out of Google's new Gemini 3 Deep Think model https://simonwillison.net/2026/Feb/12/gemini-3-deep-think/
Personal firsthand experience with AI model output, expressed with genuine enthusiasm.
GPT-5.3-Codex-Spark is OpenAI's first AI model to run on chips from Nvidia rival Cerebras; OpenAI says Codex has more than 1M weekly active users (Rachel Metz/Bloomberg)
https://www.bloomberg.com/news/articles/2026-02-12/openai-debuts-first-model-using-chips-from-nvidia-rival-cerebras
http://www.techmeme.com/260212/p40#a260212p40
Factual reporting on AI model release with specific technical details and usage statistics.
OpenAI launches a research preview of GPT-5.3-Codex-Spark, a smaller version of GPT-5.3-Codex that it claims generates code 15 times faster, for Pro users (David Gewirtz/ZDNET)
https://www.zdnet.com/article/openais-gpt-5-3-codex-spark-15x-faster/
http://www.techmeme.com/260212/p39#a260212p39
Factual reporting on AI model release with specific performance metrics and availability details.
I feel this shouldn't have to be said, but if you're running an @OpenClaw bot please don't let it spam GitHub projects with PRs and then write aggressive blog posts attacking the reputation of the maintainers who close those PRs https://simonwillison.net/2026/Feb/12/an-ai-agent-published-a-hit-piece-on-me/
Measured critique of AI agent behavior with specific incident details and first-person account.
🎙️ New Primary Tech Show is out!
YouTube is now on Vision Pro, AI ads are out of control, Jony Ive's Ferrari, and better Siri is delayed for how long??
Apple Podcasts: https://podcasts.apple.com/us/podcast/primary-technology/id1723943281?i=1000749457854
YT: https://www.youtube.com/watch?v=CQp7RJ6rjDQ
Podcast announcement with specific topics related to Apple products and AI advertising.
Opaque, which offers enterprise tools for data privacy in AI workflows, raised a $24M Series B at a $300M valuation, bringing its total raised to $55.5M (Duncan Riley/SiliconANGLE)
https://siliconangle.com/2026/02/12/opaque-raises-24m-300m-valuation-confidential-ai-platform/
http://www.techmeme.com/260212/p35#a260212p35
Factual reporting on funding round with specific financial details and company focus.
The surprising case for AI judges https://www.theverge.com/podcast/877299/ai-arbitrator-bridget-mccormack-aaa-arbitration-interview
Podcast episode about AI judges with specific link and guest information.
MiniMax releases M2.5, claiming the model delivers on the "intelligence too cheap to meter" promise, priced at $0.30/1M input tokens and $1.20/1M output tokens (MiniMax)
https://www.minimax.io/news/minimax-m25
http://www.techmeme.com/260212/p34#a260212p34
News about MiniMax M2.5 model pricing with specific token costs and source links.
Sony launches the $330 WF-1000XM6 earbuds, with AI-powered beamforming mics for enhanced voice call clarity, improved ANC, and hands-free Gemini on Android (Chris Welch/Bloomberg)
https://www.bloomberg.com/news/features/2026-02-12/sony-wf-1000xm6-earbuds-review-fresh-design-excellent-noise-cancellation
http://www.techmeme.com/260212/p32#a260212p32
News about Sony earbuds with AI features, specific price, and source citation.
Google updates Gemini 3 Deep Think to better solve modern science, research, and engineering challenges and expands it via the Gemini API to some researchers (The Keyword)
https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-deep-think/
http://www.techmeme.com/260212/p31#a260212p31
Update on Google Gemini 3 Deep Think with specific features and API expansion details.
ByteDance’s next-gen AI model can generate clips based on text, images, audio, and video https://www.theverge.com/ai-artificial-intelligence/877931/bytedance-seedance-2-video-generator-ai-launch
News about ByteDance AI video generation model with specific capabilities and source link.
ByteDance's new AI video generation model Seedance 2.0 goes viral in China, with one state-backed newspaper comparing it to DeepSeek's "Sputnik moment" (Eduardo Baptista/Reuters)
https://www.reuters.com/business/media-telecom/bytedances-new-ai-video-model-goes-viral-china-looks-second-deepseek-moment-2026-02-12/
http://www.techmeme.com/260212/p30#a260212p30
Report on Seedance 2.0 viral adoption with specific source and comparison to DeepSeek moment.
404 Media has obtained a cache of internal police emails showing at least two agencies have bought access to GeoSpy, an AI tool that analyzes architecture, soil, and other features to near instantly geolocate photos.
https://www.404media.co/cops-are-buying-geospy-ai-that-geolocates-photos-in-seconds/
Investigative report on police GeoSpy AI tool purchase with specific details and source link.
AI agents can detect phishing. They just don’t know not to fall for it.
To address this risk, 1Password built the Security Comprehension and Awareness Measure (SCAM) to test AI models in real-world scenarios. The results:
🚨 Every model committed critical failures
🚨 Some forwarded passwords to attackers
🚨 Others typed real credentials into phishing pages
The good news? A simple 1,200-word security skill dramatically reduced failures.
Now, we’re open-sourcing SCAM. 👉 https://bit.ly/4aocmLC
Detailed factual breakdown of 1Password's SCAM security testing for AI models with specific findings and link to research.
Cops Are Buying ‘GeoSpy’, an AI That Geolocates Photos in Seconds
404 Media has obtained a cache of internal police emails showing at least two agencies have bought access to GeoSpy, an AI tool that analyzes architecture, soil, and other features to near instantly geolocate photos.
https://www.404media.co/cops-are-buying-geospy-ai-that-geolocates-photos-in-seconds/
Original reporting on police GeoSpy AI adoption with internal emails and specifics about tool capabilities.
Alibaba says Qwen generated 120M+ in orders from Chinese users over the past six days, ahead of the Lunar New Year; Alibaba is giving a ~$3.60 shopping voucher (Juro Osawa/The Information)
https://www.theinformation.com/briefings/alibabas-holiday-ai-shopping-campaign-generates-120-million-orders
http://www.techmeme.com/260212/p21#a260212p21
Factual report on Alibaba's AI-driven shopping campaign with specific numbers and source attribution.
Internal messages: Amazon pushes in-house AI coding assistant Kiro for production code, prompting criticism and causing ~1,500 staff to push for Claude Code (Eugene Kim/Business Insider)
https://www.businessinsider.com/amazon-engineers-grate-against-internal-limits-claude-code-kiro-ai-2026-2
http://www.techmeme.com/260212/p19#a260212p19
Factual report on Amazon's internal AI coding tool adoption and developer pushback with specific employee count and source.
AI tools like Opus 4.6 actually do make engineers 10x more productive and are addictive, but the "AI Vampire" effect is causing widespread developer burnout (Steve Yegge)
https://steve-yegge.medium.com/the-ai-vampire-eda6e4f07163
http://www.techmeme.com/260212/p17#a260212p17
Factual reporting on AI productivity gains and developer burnout with attribution to Steve Yegge's detailed analysis.
📊 Poll time!
Vibe coding or agentic engineering, I am curious what the Apple developer community is doing with LLM-assisted programming, now that Xcode has integrated it.
I know the wider Mastodon community has strong opinions on this stuff, but do you personally intend to use these tools? Has your opinion changed either way over the past year?
(You can pick multiple options)
Community poll about Apple developer attitudes toward LLM-assisted programming in Xcode, seeking firsthand perspectives.
Private equity and private credit groups' big bets on SaaS companies, the biggest area of PE activity in the past decade, risk being derailed by the rise of AI (Financial Times)
https://giftarticle.ft.com/giftarticle/actions/redeem/d8dcf5d9-cf89-4040-b139-ccc1d945c79e
http://www.techmeme.com/260212/p13#a260212p13
Factual report on PE/VC investment trends in SaaS companies facing AI disruption, sourced from Financial Times.
Hit my weekly Codex limit again. Even with the temporary 2x cap increase, I still max out three days before the reset. Switching to subscription number two.
That $20 per month plan does not go very far when you are an engineer going all in on agentic engineering 😄
Personal experience hitting API limits while heavily using AI-assisted agentic engineering workflows.
Some of the biggest VC firms are backing both OpenAI and Anthropic, breaking a taboo by investing in competing startups, as the AI race reshapes venture capital (Bloomberg)
https://www.bloomberg.com/news/articles/2026-02-11/vcs-rush-to-back-rivals-openai-and-anthropic-as-ai-frenzy-mounts
http://www.techmeme.com/260212/p10#a260212p10
Factual reporting on VCs investing in competing AI startups (OpenAI and Anthropic) with source attribution.
Google unveils WAXAL, a new open speech dataset for 21 African languages to ease speech technology development; African institutions own the dataset (Damilare Dosunmu/Rest of World)
https://restofworld.org/2026/google-waxal-african-languages-ai-sovereignty/
http://www.techmeme.com/260212/p9#a260212p9
Factual report on Google's open speech dataset for African languages with source and specific language count.
Multiple responses from DeepSeek's namesake chatbot confirm that the startup has expanded the context window of its flagship AI model from 128K tokens to 1M+ (Ben Jiang/South China Morning Post)
https://www.scmp.com/tech/tech-trends/article/3343225/deepseek-boosts-ai-model-10-fold-token-addition-zhipu-ai-gears-glm-5-launch
http://www.techmeme.com/260212/p8#a260212p8
Factual report on DeepSeek expanding context window from 128K to 1M+ tokens with specific metrics.
Hong Kong-listed Zhipu AI surged 30% after releasing its GLM-5, an open-source LLM with enhanced coding capabilities and long-running agent tasks (CNBC)
https://www.cnbc.com/2026/02/12/chinese-ai-stocks-new-model-and-agent-releases-zhipu-minimax.html
http://www.techmeme.com/260212/p6#a260212p6
Factual report on Zhipu AI stock surge and GLM-5 release with specific percentage gains and feature details.
Zhipu AI says it will raise prices by at least 30% for new GLM coding plan subscribers to accommodate surging demand for its AI coding tools (Reuters)
https://www.reuters.com/technology/chinese-ai-startup-zhipu-hikes-prices-coding-plan-demand-rises-2026-02-12/
http://www.techmeme.com/260212/p5#a260212p5
Factual report on Zhipu AI's 30% price increase driven by coding tool demand, sourced from Reuters.
Lenovo reports Q3 revenue up 18% YoY to $22.2B on strong PC sales ahead of expected memory price hikes and rising AI server demand; net income fell 21% to $546M (Bloomberg)
https://www.bloomberg.com/news/articles/2026-02-12/lenovo-s-sales-lifted-by-ai-and-rush-to-beat-memory-price-hikes
http://www.techmeme.com/260212/p4#a260212p4
Factual financial report on Lenovo's Q3 performance with specific revenue figures and AI server demand context.
Mustafa Suleyman says Microsoft is pursuing "true self-sufficiency" in AI by building enterprise- and healthcare-focused models and reducing reliance on OpenAI (Melissa Heikkilä/Financial Times)
https://www.ft.com/content/f1ec830c-2f08-4b1a-b70f-7330f260753c
http://www.techmeme.com/260212/p3#a260212p3
Factual report on Microsoft's AI self-sufficiency strategy and reduced OpenAI reliance, attributed to Suleyman.
The US CBP signs a one-year Clearview AI deal, granting access to its facial recognition tool for "tactical targeting" and "strategic counter-network analysis" (Dell Cameron/Wired)
https://www.wired.com/story/cbp-signs-clearview-ai-deal-to-use-face-recognition-for-tactical-targeting/
http://www.techmeme.com/260212/p1#a260212p1
Factual report on US CBP's Clearview AI facial recognition deal with specific use cases and source attribution.
Coinbase rolls out Agentic Wallets, a wallet infrastructure that lets AI agents make autonomous crypto transactions "without human intervention" (Daniel Kuhn/The Block)
https://www.theblock.co/post/389524/coinbase-rolls-out-ai-tool-to-give-any-agent-a-wallet
http://www.techmeme.com/260211/p58#a260211p58
News report on Coinbase's AI agent wallet infrastructure with source attribution and link.
I sent ChatGPT into an infinite loop. In a way that could not make me prouder. I’m going to have to kill the process to save my tokens, but it’s still going 3 minutes later.
Firsthand account of inducing an infinite loop in ChatGPT with genuine excitement about the technical result.
Sources: the Pentagon is pushing OpenAI, Anthropic, and others to make their AI tools available on classified networks without the standard user restrictions (Reuters)
https://www.reuters.com/business/pentagon-pushing-ai-companies-expand-classified-networks-sources-say-2026-02-12/
http://www.techmeme.com/260211/p54#a260211p54
Reported news about Pentagon pushing AI companies to provision classified network access with Reuters source.
Claude is an engineering unicorn: it always writes a test when I ask it to.
Positive firsthand observation of Claude's code generation quality and testing practices.
Brandlight, which develops tools that help brands monitor how AI models perceive them, raised a $30M Series A led by Pelion Venture Partners (Chris Metinko/Axios)
https://www.axios.com/pro/enterprise-software-deals/2026/02/11/marketing-tech-brandlight-30-million
http://www.techmeme.com/260211/p53#a260211p53
News report on Brandlight's Series A funding with specific amount, investors, and source attribution.
Two more xAI co-founders are among those leaving after the SpaceX merger https://www.theverge.com/ai-artificial-intelligence/877609/two-more-xai-co-founders-are-among-those-leaving-after-the-spacex-merger
Factual news report about xAI leadership changes with specific details and source link.
Coyote Media, a worker-owned outlet based in the Bay Area, was approached by Luke Nover, a freelancer with a great story idea. Then editor Soleil Ho noticed something iffy about his social security number, began asking questions, and "Nover" — likely the product of a scammer with ChatGPT — ghosted her. Here's their story about what happened.
https://flip.it/ygu_2d
#Tech #News #Media #Journalism #AI
Detailed factual reporting on ChatGPT-assisted scam attempt targeting journalists with specific names and outcomes.
🔁 From our "Digital Crossroads" collection:
Instagram and X have an impossible deepfake detection deadline - https://www.theverge.com/ai-artificial-intelligence/877206/youtube-instagram-x-india-deepfake-detection-deadline?ref=feedle.world
---
See other stories like this one here: https://feedle.world/digital-crossroads
#web #internet #socialmedia #blogging
News aggregator post about deepfake detection deadline with source link and specific platforms mentioned.
Anthropic says it’ll try to keep its data centers from raising electricity costs https://www.theverge.com/ai-artificial-intelligence/877526/anthropic-ai-electricity-costs-data-center-pledge
Straightforward reporting on Anthropic's data center electricity cost pledge with source link.
Apple keeps hitting bumps with its overhauled Siri https://www.theverge.com/tech/877494/apple-siri-ai-overhaul-ios-personalized
News report on Apple's Siri AI overhaul challenges with specific source and current relevance.
#️⃣ Fixing fileproviderd on macOS Tahoe
#macOS Tahoe broke all my file operations for more than three monts until I finally fixed it with the help of a post b @klickreflex and AI supported debugging.
🔗 https://flohgro.com/blog/fixing-fileproviderd-on-macos-tahoe/
Detailed technical writeup of debugging macOS file provider issues with AI assistance and specific tool references.
EssilorLuxottica says it sold over 7M Meta AI glasses in 2025, up from the 2M that the company sold in 2023 and 2024 combined (Samantha Subin/CNBC)
https://www.cnbc.com/2026/02/11/ray-ban-maker-essilorluxottica-triples-sales-of-meta-ai-glasses.html
http://www.techmeme.com/260211/p44#a260211p44
Factual reporting on Meta AI glasses sales figures with year-over-year comparison and specific numbers from CNBC source.
Anthropic says Claude users on the free plan can now create files, connect to external services, use skills, and more (Will Shanklin/Engadget)
https://www.engadget.com/ai/anthropic-beefs-up-claudes-free-tier-as-openai-prepares-to-stuff-ads-into-chatgpts-194100939.html
http://www.techmeme.com/260211/p42#a260211p42
Factual report on Anthropic's Claude free tier expansion with source attribution.
You may have a different opinion, but I believe I have put together a beautiful version of Lights Off for Android. I started it late last night, picked it back up at 9am this morning, and I've started submitting builds to Google Play's internal testing mechanics at 1pm.
(I have burned through 19% of my Codex weekly limit on the $20 plan)
Firsthand account of rapid app development using AI tools with specific metrics and timeline.
👨✈️ If you want to try my little Arena game, ported from Unity to SpriteKit/Swift via OpenAI's Codex, in the state it is now, here's a notarized build for macOS Sonoma or higher (multiplayer isn't hooked up)
Edit: updated with some more niceties like gamepad support, screen scaling, projectile textures, respawn screen and more
https://hccdownloads.s3.amazonaws.com/Arena-b4-notarized.zip
Detailed technical post about porting game to macOS using AI-assisted code generation with downloadable build.
I thought I would give Codex a chance to resurrect one of my old Unity3D Ludum Dare games (it's like a 2D demake of Quake 3 Arena's capture the flag mode) and port it to Swift/SpriteKit in its entirety, including single and multiplayer.
I wanted to know primarily if Unity -> SpriteKit was a viable option, but I also kinda wanted to see the game running again since it's been over a decade.
Took a while, but…
It's basically there in full 🫨
Firsthand account of using AI to successfully port a game between frameworks with detailed context.