Workplace, Policy, Misuse
· The Fluency Briefing
The Fluency Briefing
Your Guide to What's Happening in AI and Why It Matters to You
Wednesday, March 11, 2026
Three stories landed today that, taken together, tell one story: the gap between AI ambition and AI reality is where all the action is. Anthropic is launching a think tank while suing the Pentagon, Amazon employees say AI tools are creating more work than they save, and Accenture's CEO just made AI skills a prerequisite for promotion. The pattern?
Everyone agrees AI matters - nobody agrees on what to actually do about it.
Today in AI:
- Amazon's AI Tools Are Making More Work, Not Less - Corporate employees told the Guardian that Amazon's internal AI coding tool Kiro frequently hallucinates, forcing developers to spend time fixing AI-generated code rather than writing their own. One developer described it as "trying to AI my way out of a problem that AI caused." The Guardian
- Accenture: No AI Skills, No Promotion - Accenture CEO Julie Sweet confirmed that AI proficiency is now required for advancement at the consulting giant, while also doubling down on entry-level hiring. The move signals that large employers are betting on retraining over replacing. Fast Company
- Anthropic Launches Think Tank While Battling the Pentagon - Anthropic announced the Anthropic Institute, an internal research group studying AI's societal impacts, and is tripling its DC policy team. This comes days after the company sued the US government over a Defense Department blacklist. The Verge
- Teens Are Using AI to Roast Their Teachers on TikTok - Student-run "slander pages" are using AI video tools like Viggle AI to create deepfake memes mocking school faculty, some comparing teachers to notorious public figures. One post racked up over 107,000 likes. Wired
- Google Brings Gemini in Chrome to India, Canada, and New Zealand - The expansion adds a sidebar AI assistant to Chrome on desktop, supporting eight Indian languages and letting users summarize pages, compare tabs, and interact with Gmail and YouTube without leaving the browser. TechCrunch
- China's OpenClaw Craze Is Minting a New Class of AI Hustlers - The viral "raise a lobster" AI trend has spawned packed meetups of 1,000-plus attendees in Shenzhen and local government subsidies. People with technical skills are charging to help non-technical users install the platform, turning setup into a side business. MIT Technology Review
- Meta Deploys AI to Fight the Scams AI Helped Create - Meta is rolling out AI-powered tools to detect brand impersonators, deepfake celebrity scams, and deceptive links across Facebook, WhatsApp, and Messenger. The company removed 159 million scam ads and 10.9 million scam accounts throughout 2025. Engadget
- Khosla Ventures Partner Says Entry-Level Jobs Are Disappearing - Ethan Choi told Crunchbase News he's watching AI replace the research and analysis tasks that once defined junior roles, calling it a shift in the social contract of the modern workforce. He now uses voice-mode AI to prep for meetings while driving. Crunchbase News
Today's Takeaway:
Here's the thing about Amazon's AI stumble: it's not just an Amazon problem, it's a preview of what happens when companies mandate AI adoption faster than the tools can deliver. According to The Guardian, corporate employees describe a culture where using AI isn't optional, even when the output is worse than doing it manually. Developers spend their days cleaning up hallucinated code from Kiro, Amazon's internal tool. Translation: the company built to obsess over efficiency is creating inefficiency - and requiring employees to participate in it.
Now contrast that with Accenture's approach. As Fast Company reports, CEO Julie Sweet is tying promotions to AI proficiency but still investing in entry-level hiring and what she calls "leader-led learning." The difference matters. Amazon is pushing tools down from the top and hoping people adapt. Accenture is pushing skills up from the workforce and betting on humans getting better at using AI, not just being subjected to it. One company treats AI as a mandate; the other treats it as a capability to develop. If you're watching your own employer roll out AI tools right now, which approach looks more familiar?
💡 Fluency Moment - Building your AI fluency, one term at a time.
"Hallucination"
In plain English: When an AI confidently produces false or made-up information as if it were true.
Think of it like: A new intern who answers every question with total confidence, even when they're completely making things up.
Why you'll hear about it: Amazon developers wasted hours fixing AI-generated code full of hallucinated errors.
🧰 Your Toolkit
5-Minute Quickstart: Using AI in Your Everyday Life
- Open ChatGPT or Google Gemini in your browser - no account needed to try Gemini at gemini.google.com.
- Type a simple question like 'Explain [topic you're curious about] as if I'm 10 years old' and read the response.
- Ask AI to help with a real task: 'Write a short email to [person] about [topic] in a friendly tone.'
- Try Gemini in Chrome by clicking the sidebar icon and asking it to summarize the webpage you're currently reading.
- Ask AI for advice on something you're learning: 'Give me 3 beginner tips for [hobby or skill you want to try].'
- Save a useful response by copying it into a notes app so you can refer back to it anytime.
Once you're comfortable with basic questions, try asking AI to help you plan a goal or solve a problem at work or school. The more specific your question, the more useful the answer!
The Bottom Line
The Pattern: Across every story today - Amazon's coding tool mess, Accenture's promotion requirements, Anthropic's policy scramble, teens deepfaking their teachers - the same tension keeps surfacing. AI is being deployed faster than anyone has figured out the rules, norms, or quality bars for using it well.
Why It Matters: The companies and individuals who learn to distinguish between "AI that actually helps" and "AI deployed for the sake of deploying AI" will have a meaningful edge over the next two years. Khosla's Ethan Choi is right that entry-level jobs are shifting, and Accenture is right that AI skills now matter for career advancement. But Amazon's experience is proof that mandating bad tools erodes trust and slows people down - the opposite of the point.
Your Move: This week, audit one AI tool you currently use at work. Time how long you spend fixing or working around its output versus how long the task would take without it. If AI is costing you time, that's data worth bringing to your manager - before someone decides to double down on it anyway.
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