• Everyday AI
  • Posts
  • Ep 715: OpenAI's OpenClaw Acquisition And Anthropic's Disastrous 2026

Ep 715: OpenAI's OpenClaw Acquisition And Anthropic's Disastrous 2026

Grok 4.20 Released, Claude's new Sonnet model, Figma Teams Up with Anthropic for AI-Powered Design Tool and more

Author

Everyday AI
February 17, 2026

Sup y’all! 👋

For tomorrow’s show, we’re diving deep into ChatGPT’s new (and crazy useful!) new Deep Research update.

Don’t miss it… but wondering what your current Deep Research fave is?

What's your favorite Deep Research tool?

🗳️ Vote to see LIVE results 🗳️

Login or Subscribe to participate in polls.

See ya tomorrow.

✌️

Jordan

P.S. — let’s connect on LinkedIn! Just tell me you’re from the newsletter cuz stranger danger.

Outsmart The Future

Today in Everyday AI
8 minute read

🎙 Daily Podcast Episode: OpenAI’s OpenClaw acquihire signals a major shift in the AI agent race. The developer battle between OpenAI and Anthropic just got real. Give today’s show a watch/read/listen.

🕵️‍♂️ Fresh Finds: Unitree’s humanoid robots put on a kung fu performance, Meta’s Ads Manager now includes Manus AI tools, Google Stitch is testing its new “Hatter” agent and more. Read on for Fresh Finds.

🗞 Byte Sized Daily AI News: Grok 4.20 Released, Claude's new Sonnet model, Figma Teams Up with Anthropic for AI-Powered Design Tool and more. Read on for Byte Sized News.

💪 Leverage AI: OpenAI just acqui-hired the creator of OpenClaw, the most viral AI agent on the internet. Anthropic chose lawyers over partnership, and that decision may have reshaped the agent ecosystem. Keep reading for that!

↩️ Don’t miss out: Miss our last newsletter? We covered: OpenAI acquires OpenClaw, Meta brings Manus AI agents to Telegram, Amazon prepares to spend $200 billion expanding AWS data centers and more. Check it here!

Ep 715: OpenAI's OpenClaw Acquisition And Anthropic's Disastrous 2026

Yeah, you prolly saw the news: OpenAI acquihired OpenClaw. 🦞

But what you probably didn't see/read/notice: this will completely change the agent landscape and Anthropic's fumbling of the original Clawdbot could be a misstep that costs them billlllllions.

Also on the pod today:

• OpenClaw's viral GitHub takeover 🦞 
• $16M crypto scam hijack 💸 
• Anthropic's lawyer move flop ⚖️

It’ll be worth your 35 minutes:

Listen on our site:

Click to listen

Subscribe and listen on your favorite podcast platform

Listen on:

Here’s our favorite AI finds from across the web:

New AI Tool Spotlight – Figr is Product-aware AI that thinks through UX, then builds it, HostedClaws is Your own AI employee, Vela is your AI scheduling assistant that gives you back your most valuable resource: time.

Social Media Posts After Death — Meta patented AI that could mimic you online after death. It's just a concept, but it's raising big questions.

AI Accusations — A former Morning Edition host says Google used his voice for an AI product without permission. Find out what he alleges and how Google responds.

Microsoft Tests “Tasks” Feature — Microsoft is testing a unified “Tasks” feature for Copilot that lets you schedule and automate complex workflows. Want to see what it can do?

ChatGPT Web App Feature — OpenAI is testing a new feature that lets you set your default model and pick how much "thinking effort" it uses.

Unitree Gala Robots — Dozens of humanoid robots perform Kung Fu—fully autonomously.

Google Stitch Hatter Agent — Google Stitch is testing a “Hatter” agent for smarter design and automatic app store assets.

Research: Harvard on GenAI — AI adoption looks good on paper, but fears about job security are holding back real impact. Curious why?

AI Companies — 17 US AI startups have already raised $100 million or more in 2026. See which names are leading the funding frenzy.

Mistral Buys Koyeb — Mistral bought cloud startup Koyeb and is investing big in European data centers to stay local.

The Vatican AI Assisted Service — The Vatican is rolling out AI-powered live translations for Mass in 60 languages, no app needed. Scan a QR code and follow along in real time.

Boundary Point Jailbreaking — BPJ cracks even the toughest AI safety systems, revealing new gaps in AI security. Curious how?

Meta Manus Ads — Meta’s Ads Manager now features Manus AI tools. See how they might level up your campaigns.

Rwanda and Anthropic — Rwanda and Anthropic team up to bring AI to schools, healthcare, and government. Find out what’s next.

1. Grok 4.20 Beta Launches with Four-Agent Power-Up 🧠

xAI has begun rolling out Grok 4.20 (Beta), a new upgrade featuring four collaborating AI agents across all apps, according to TestingCatalog News.

This marks a fresh push in the ongoing AI model race, promising more dynamic teamwork and faster responses. While some insiders suggest the upgrade may not be fully live, excitement and skepticism are both running high as users put the new system through its paces.

2. Figma Teams Up with Anthropic for AI-Powered Design Tool ⚒️

Figma just launched “Code to Canvas,” a new feature built with Anthropic that lets users turn AI-generated code into editable designs inside Figma.

This move comes as software stocks tumble and Figma’s own shares drop 85% since its IPO, highlighting the stormy climate for SaaS companies amid rapid AI advances. The partnership bets that even as AI coding tools improve, design refinement still matters—but there’s a risk teams could eventually skip Figma’s step altogether.

3. Alibaba Unveils Qwen3.5-397B: Open-Source AI Giant Arrives 💻

Alibaba has just launched Qwen3.5-397B, the first open-weight model in its new AI series, making a big splash in the race for smarter, faster language models.

This next-gen system is natively multimodal and tailored for real-world use, boasting huge speed gains and support for 201 languages and dialects. With its hybrid attention architecture and large-scale RL training, Qwen3.5-397B promises to out-decode competitors by a wide margin.

4. Anthropic and Infosys Join Forces on Enterprise AI Expansion in India 🤝

Anthropic and Infosys have just unveiled a new partnership to bring advanced AI solutions to major industries in India, including telecommunications, finance, and manufacturing.

By combining Anthropic’s Claude models with Infosys’ Topaz platform, the collaboration aims to speed up software development and ensure responsible AI adoption in regulated sectors. The focus is on building AI agents that can handle complex tasks like compliance and system modernization without human intervention.

5. India Sets Sights on $200B AI Data Center Boom 🤑

India is making a major play to become the world’s next AI heavyweight, targeting up to $200 billion in data center investments as global tech giants like Google, Microsoft, and Amazon ramp up their spending.

The government is betting big on homegrown AI talent, digital infrastructure, and accessible computing power, aiming to make artificial intelligence a tool for the masses, not just the elite. With new tax breaks, a nationwide GPU facility, and a push for sovereign AI models in local languages, officials say India wants to move beyond being just a market for tech—it wants to write the rules.

Please summarize this {PAGE} in an engaging and insightful way, as if you were a news reporter covering AI news. Ensure the news-focused recap is very short and only 4 sentences max, focusing on the timeliness of the news event or announcement. start the summary with a very short bolded headline with one emoji at the end. then, continue the summary with a new paragraph, keeping it only 4 sentences long total. make sure to respond in an engaging and witty tone, yet keep it confident, factual and non-confrontational. It should read like a news report, but slightly more engaging and entertaining. It should NOT read like a press release. Please do not make the responses cheesy or dorky. Make sure to highlight the main points of the {PAGE}, and always put it into context and explain what the content on the {PAGE} means in simple terms without giving examples. Make sure to first highlight the timeliness or newsworthiness of the information on the {PAGE}, as that should always lead the news story. Cite the source of the article or the study only if it's relevant or an exclusive report, "According to (source name) or A new study from (source) says that ..." Avoid using em dashes (—) and use both medium length and long sentences. Shoot for simple, straightforward recaps that include details and context, yet avoid jargon and confusing stats. Focus on big picture newsworthiness and straightforward context.

6. Anthropic Ups the AI Ante with Claude Sonnet 4.6 🆕

Anthropic has just released Claude Sonnet 4.6, marking its second major AI model upgrade in less than two weeks and setting a rapid pace in the ongoing AI race. The new model now powers both free and paid Claude users, promising sharper coding, design, and data analysis skills that rival top-tier AI offerings.

With this launch, Anthropic aims to make advanced AI productivity accessible and shake up the competitive landscape dominated by OpenAI and Google.

OpenAI just acqui-hired the creator of the most popular AI agent, OpenClaw. The same open source project that was probably making Anthropic millions.

How did Anthropic fumble? 

They sent lawyers instead of a partnership offer.

(And TBH…. this epic fumble may have decided who controls the agent era this year.)

We unpacked the full story on today's Everyday AI Hot Take Tuesday and laid out exactly how the OpenClaw acqui-hire reshapes which platform your future agents will likely run on.

If you're betting your agent strategy on the wrong ecosystem, Monday morning needs a new conversation.

Here’s what ya need to know.

1. Anthropic Chose Lawyers Over Community 🔥

OpenClaw started as Clawdbot. Built on Anthropic's API by default. And it took off, going mega viral in early 2026 and very likely bringing with it a massive amount of new and increased Anthropic subscriptions.

Developers were signing up for Anthropic licenses by the thousands just to use it. Its founder Peter Steinberger was running a free marketing machine that Anthropic couldn't have bought for 8 figures. 

Then Anthropic's legal team told him to rename it. Because C-L-A-W-D was too close to C-L-A-U-D-E.

Sure, we get the whole protecting the brand thing. But cmon! How did not a single adult in the room understand the implications of being penny smart, dollar dumb? 

So Clawdbot molted as Moltbot for like 48 hours, eventually going through a final transformation and became OpenClaw. Steinberger started building and sharing about how he was using OpenAI's Codex. Sam Altman welcomed the name, backed the project, and completed the acqui-hire on Sunday.

OpenClaw transitions to an OpenAI-funded foundation and stays open source. But the default optimization is prolly shifting to GPT models. (Thought we assume OpenClaw’s open source nature will obviously allow you to power it with whatever model you want) 

Think Chrome and Chromium.

Anthropic could have captured the community, the tech, the momentum, the relationship and the likely massive revenue that came along with it. They chose cease-and-desist letters over a handshake. Yikes. 

Try This

Pull up every open source AI tool your team uses this week and check who actually controls governance and model defaults.

If any critical workflow defaults to a single provider, that's a strategic dependency. Not a feature.

Developer loyalty shifts faster than vendor contracts right now. Don't find out your favorite tool switched ecosystems from a tweet.

2. The Developer Exodus Has Receipts ⚡

Here’s the receipts we dug up that no one is talking about. According to OpenRouter data, Anthropic held 40% API market share a year ago.

This week? Below 10%.

Not even top three.

That ain't a dip. That's a freefall y’all. 

And it's not just one thing. OpenAI's Codex platform is shipping at a ridiculous pace. The new Codex Spark model is hitting 1,000 tokens per second. Chinese open source models are getting scary capable. Developers are voting with their tokens and Anthropic is no longer the defacto dev choice. 

Stack that with Anthropic's Super Bowl ad ranking in the bottom three percent for likability over five years according to iSpot. Only seven percent of viewers even knew what Claude was.

(Sorry…. but running an attack ad when nobody knows your name is like trash-talking at a party where nobody invited you.)

Then we got word this week that the Pentagon is reportedly threatening to label Anthropic as ‘a supply chain risk’ over ethics restrictions. Plus a $1.5 billion copyright settlement from authors. 

Anthropic is shipping its best products ever while somehow losing the narrative literally everywhere. 

Yikes again. 

Try This

Run a real comparison this week. Take your three most common AI workflows and test them on your current provider plus two alternatives.

Not benchmarks. Your actual Monday morning tasks.

The provider winning developer hearts today is prolly building the best agent infrastructure tomorrow. Check OpenRouter's public rankings to see real usage versus marketing.

3. The Agent Era Kingmaker Just Shifted 🚀

Both Anthropic and OpenAI are heading toward IPOs this year, but first to market and momentum means everything. 

And the crazy thing like a  month ago, OpenAI’s lead in those areas was seemingly slipping. 

Then Anthropic started dropping balls and tarnishing its own rep. 

OpenAI? Making huge dev moves, the area that Anthropic has been dominating. 

OpenAI now has Codex CLI (open source), OpenClaw (open source), and their GPT-OSS models. They went from being called "Closed AI" to straight up owning the open source agent narrative.

Meanwhile, Anthropic's CEO Dario Amodei is out there warning that AI could take 50% of white collar jobs. Respect the honesty, for sure. But that ain't exactly a brand message that makes enterprises wanna double down on your platform.

Real talk.

A month ago, it felt like Anthropic could catch OpenAI this decade. Claude Code, their new plugins, Claude Cowork, and Opus 4.6 were all straight up ON FIRE! NBA Jam style. 

But when you send cease-and-desist letters to your biggest fans, flop a Super Bowl ad thinking you’re bigger than you are, and fight with the Pentagon….. AND start to lose your developer base in the same month? 

That belief is gone.

Triple yikes. 

Try This

If you're evaluating AI platforms for enterprise agent deployment, watch ecosystem health more than benchmark scores.

The provider with the strongest open source community prolly ends up owning the agent infrastructure long term.

Platform decisions made this quarter will compound hard. Choose based on where developers are going, not where they were.

Reply

or to participate.