AI: Regulation, Ads, Automation
· The Fluency Briefing
The Fluency Briefing
Your Guide to What's Happening in AI and Why It Matters to You
Monday, February 9, 2026
Everyone is chasing the 'best' AI, but what if the race itself is rigged?
A new MIT study reveals the leaderboards we rely on are surprisingly fragile, while regulators and rivals are busy trying to redraw the map entirely. It seems the era of blindly trusting the AI hype is officially over.
Today in AI:
- The EU Wants a Word with Meta - The European Commission is telling Meta it can't block rival AI chatbots from WhatsApp. The EU argues that locking out competitors like ChatGPT from a major messaging app is an abuse of power, setting the stage for another Big Tech vs. Brussels showdown. Bbc
- AI Leaderboards Are Surprisingly Fragile - Think you know which LLM is number one? An MIT study found that removing just a tiny fraction of user feedback data can completely change which models top the charts, warning businesses not to blindly trust the rankings when making costly decisions. MIT News
- Your ChatGPT Convos Might Get Ads - OpenAI is reportedly starting to test ads in the free version of ChatGPT. While the company says ads will be a minority of its revenue long-term, it's a clear sign that the 'free AI' party is winding down as the search for profits heats up. Cbsnews
- AI Fears Tank Software Stock - Project management tool Monday.com saw its stock plummet 19% after issuing weak guidance. The drop reflects growing investor anxiety that new, powerful AI agents could simply replace existing software-as-a-service platforms, rattling the entire industry. cnbc.com
- Siemens' Plan to Automate Everything - Siemens CEO Roland Busch laid out a vision to automate not just factory floors, but also the digital work that supports them, like accounting and procurement. It's a glimpse into a future where AI manages the entire business process, for better or worse. The Verge
- Anthropic Upgrades Its Coding Sidekick - Anthropic just dropped Claude 4.6 Opus, a major update focused on 'agentic coding.' Translation: It's getting much better at taking a natural language request and turning it into functional code, further blurring the line between human and AI developers. whytryai.com
- AI Still Thinks My Wife Is My Colleague - A columnist for The Guardian played a simple game: 'Ask AI who my wife is.' The AI confidently and incorrectly named several authors, a newsreader, and a colleague, providing a hilarious and humbling reminder that these systems are still prone to making things up. Dailymail Co Uk
Today's Takeaway:
We all love a good leaderboard. It’s simple, definitive, and tells us who’s winning. But when it comes to large language models, those rankings might be built on a house of cards. A new study from MIT researchers reveals that the popular platforms ranking LLMs are incredibly sensitive to small changes in data. By removing just a handful of user votes from a dataset of tens of thousands, the researchers could completely reshuffle the top-ranked models. This isn't just an academic curiosity; it's a major red flag for any business using these leaderboards to make six-figure decisions about which AI to integrate into their products. MIT News
Here’s the thing: these platforms rely on crowdsourced feedback, where users vote on which of two anonymous AI responses is better. The study shows that a few 'influential votes' can have an outsized impact, skewing the results and creating a false sense of certainty. It suggests that what we perceive as a clear winner might just be the model that got lucky with a few specific prompts. The takeaway isn't that rankings are useless, but that they should be taken with a huge grain of salt. For anyone choosing an AI tool, the only leaderboard that truly matters is the one you create by testing models on your own specific tasks.
💡 Fluency Moment - Building your AI fluency, one term at a time.
"Context Window"
In plain English: The amount of text an AI can "remember" during a single conversation.
Think of it like: A notepad with limited pages - when it fills up, the AI forgets earlier notes.
Why you'll hear about it: Bigger context windows mean AI can handle longer documents and conversations.
🧰 Your Toolkit
5-Minute Quickstart: Trying Your First AI Tools
- Sign up for a free AI chatbot like ChatGPT or Google Gemini and ask it a fun question, like 'What's a good joke about computers?'
- Now, ask it to help with a daily task. Try this prompt: 'Write a short, friendly email to [Person's Name] about [Topic].'
- Visit a free AI image generator like Microsoft Designer and describe a simple scene, such as 'A happy robot drinking coffee on the moon.'
- Give the same creative prompt to two different AI chatbots. For example, ask both to write a short poem about [Your Favorite Hobby].
- Compare the results from the two tools. Notice the differences in style and decide which one you prefer for that specific task.
Now that you've tried the basics, think of one task you do every week and see how AI can help you brainstorm or draft it.
The Bottom Line
The Pattern AI is having its awkward teenage years. It's moving out of the research lab and into the real world, where it has to get a job (OpenAI's ads), follow house rules (EU vs. Meta), and deal with the fact that it's not as smart as it thinks it is (MIT's study).
Why It Matters This is the messy but necessary transition from hype to utility. The winners won't be chosen by abstract benchmarks, but by real-world performance, sustainable business models, and their ability to navigate a complex regulatory landscape.
Your Move Stop asking 'Which AI is best?' and start asking 'Which AI is best for this specific task?' Test two different models on a real work problem this week and see for yourself.
What We're Working On
✨ Founding Cohort Special - 60% Off! - Use code MAF20 to join for just $20/month (regularly $50). Get weekly group sessions & workshops, self-paced courses for all levels, access to tools & templates, challenges with peer feedback, and 24/7 support community. → Join Now
✨ Free 30-Minute AI Consultation - Discover how My AI Fluency can help your business unlock the potential of AI. We'll discuss your goals, explore practical AI opportunities for your industry, and outline clear next steps. → Schedule Free Call
💬 Community | 📞 Book a Consultation | 🌐 Website
Fluently yours, The My AI Fluency Team