Voice, Financial AI, Jobs
· The Fluency Briefing
The Fluency Briefing
Your Guide to What's Happening in AI and Why It Matters to You
Saturday, March 28, 2026
A 230-year-old waste-sorting company is saving 40% of its knowledge-work time with ChatGPT. Mistral just open-sourced a voice AI model to challenge ElevenLabs. And Intuit wants to replace your CFO with an algorithm that processes 60 billion predictions a day. Three very different companies, one shared bet: AI isn't a side project anymore, it's becoming the operating layer.
Here's what that actually looks like on the ground this Saturday.
Today in AI:
- Mistral Hands Out Free Voices - Mistral AI released Voxtral TTS, an open-weight text-to-speech model it claims outperforms ElevenLabs. The key difference: companies can run it on their own servers without sending audio to a third party. The voice AI market is now north of $22 billion globally. VentureBeat
- Intuit Wants to Be Your AI CFO - Intuit is turning QuickBooks from a digital ledger into a real-time financial decision-maker, processing roughly 60 billion ML predictions per day. Revenue hit $4.7 billion last quarter, up 17% year over year, as AI-driven features gain traction with small businesses. Fast Company
- A 230-Year-Old Company Goes AI-First - STADLER, a family-owned recycling plant manufacturer, deployed ChatGPT across nearly every department and built over 125 custom GPTs. The result: 30-40% time savings on common knowledge tasks and 2.5x faster first drafts on average. OpenAI
- Your Job Isn't Dying, It's Mutating - A Harvard Business School study found that while postings for automatable roles dropped 13% from 2019-2025, demand for analytical and creative roles grew 20%. PwC data shows wages are rising twice as fast in AI-exposed industries. Fast Company
- Deepfakes Work Even When You Know They're Fake - Researchers found that AI-generated military images, including sexualized women in camouflage, are generating audiences and serving as effective propaganda even when viewers recognize them as fabricated. The emotional residue sticks regardless. The Guardian
- SK Hynix Eyes a Blockbuster U.S. IPO - The South Korean memory chip giant, already worth around $440 billion, confidentially filed for a U.S. listing that could raise $10-14 billion. SK hynix makes the high-bandwidth memory chips that power Nvidia's AI hardware, but trades at a discount to U.S. peers. TechCrunch
- Google Says Quantum D-Day Is 2029 - Google warned that cryptography needs to be quantum-ready within three years. Quantum computing scientist Scott Aaronson agreed the timeline is plausible. Two mathematicians just won the Turing Award for work on post-quantum cryptography models. Semafor
- Apple at 50: Still Betting on the iPhone - As Apple celebrates its 50th anniversary, hardware SVP John Ternus and marketing chief Greg Joswiak told WIRED the company plans to win the AI era while keeping the iPhone at the center. Translation: AI is the strategy, but the hardware cash cow isn't going anywhere. Wired
Today's Takeaway:
Here's the thing about this Saturday's news: the companies winning with AI right now aren't the ones making the flashiest announcements. They're the ones quietly rewiring how work actually gets done. STADLER, a recycling equipment company founded before Napoleon was in power, rolled out 125 custom GPTs and cut knowledge-work time by 30-40%, according to OpenAI. Intuit, processing 60 billion ML predictions daily, is turning accounting software from a rear-view mirror into a co-pilot, with revenue climbing 17% to prove the market is buying it, per Fast Company. Meanwhile, Mistral is betting that enterprises don't want to rent their AI voice from a cloud provider; they want to own it outright, as VentureBeat reported.
The pattern connecting these three moves is control. STADLER controls how its employees interact with AI through guardrails and custom tools, not by outsourcing to a vendor's generic product. Intuit controls the financial data pipeline end-to-end so its AI can act, not just report. Mistral is giving enterprises the weights so they control their own voice infrastructure. The companies treating AI as something they shape to fit their specific operations are pulling ahead. The ones still running disconnected pilot programs are the ones panelists at Sifted described as stuck. The difference between AI adoption and AI results is ownership of the process.
💡 Fluency Moment - Building your AI fluency, one term at a time.
"Open-Weight Model"
In plain English: An AI model whose core files are publicly released so anyone can download and run it themselves.
Think of it like: Like getting a recipe instead of just the meal - you can cook it in your own kitchen, your way.
Why you'll hear about it: Mistral just open-weighted a voice AI, letting businesses avoid sending data to outside servers.
🧰 Your Toolkit
Try This Prompt: Understanding AI's Role in Today's Biggest News Stories
Explain to me like I'm 12 years old how AI and new technology are being used in modern wars. What kinds of tools are involved, and why does it matter that they are cheaper than older weapons? I keep seeing realistic-looking fake images and videos of [a politician / a celebrity / a public figure] online. Help me make a list of simple questions I can ask myself to figure out if an image or video might be AI-generated before I share it. Imagine an AI assistant that can work on a task for hours without stopping, like a tireless intern. Explain in simple terms what that means for everyday jobs, and give me [3] examples of tasks someone like me could use it for at home or at work. The stock market dropped recently partly because of worries about war and big tech company lawsuits. Explain in plain English how scary news events can affect the value of tech companies, as if you were explaining it to someone who has never invested before. I want to understand deepfakes without feeling overwhelmed. Give me a beginner-friendly, 5-bullet summary of what deepfakes are, why people create them, and one simple thing I can do to protect myself from being fooled by them.
For the best results, paste these prompts into ChatGPT or any AI chat tool and add a detail or two about your own life to make the answers feel more personal and relevant to you.
The Bottom Line
The Pattern: Across every story today, from a 230-year-old recycling firm to a $13.8 billion French startup to Intuit's financial engine, the companies gaining ground aren't just using AI. They're embedding it into how their businesses think, speak, and make decisions. The common thread is integration over experimentation.
Why It Matters: If you're still treating AI as a tool you open in a separate tab, you're operating at a structural disadvantage against competitors who've woven it into every workflow. The Harvard data showing 20% growth in analytical roles alongside a 13% drop in automatable ones isn't a future prediction. It's already your job market.
Your Move: Pick one repeating task you did this week, drafting, summarizing, translating, categorizing, and build a specific AI workflow around it before Monday. Not a general exploration. One task, one tool, one measurable improvement. That's how a 230-year-old company got 40% of its time back.
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