Military AI Ethics Battle

2026-03-06 · The Fluency Briefing

Welcome to your weekly dose of AI wisdom, delivered fresh for March 6th, 2026! This week, we're diving deep into the latest breakthroughs, dissecting emerging trends, and offering our take on the stories shaping the future of artificial intelligence. Get ready for a curated selection of news, insightful analysis, and maybe even a prediction or two.

📰 The Big Story

There's one question that split the entire AI industry this week: Should your AI model be allowed to help wage war?

Anthropic drew a hard line, refusing to let its Claude models be used for autonomous weapons or mass surveillance — and the Pentagon responded by blacklisting them theverge.com, Feb 28. President Trump ordered all federal agencies to "immediately cease all use of Anthropic's technology" engadget.com, Mar 1. Here's where it gets interesting: reports emerged that the U.S. had already used Anthropic's AI in planning its attack on Iran, apparently after the ban was issued engadget.com, Mar 1.

The public response? Anthropic's Claude app shot to No. 1 on Apple's free app charts cnbc.com, Mar 1. Translation: consumers rewarded the company for saying no to the military.

Meanwhile, OpenAI walked straight into the mirror-image disaster. After announcing its own Department of Defense deal, ChatGPT uninstalls surged 295% in a single day techcrunch.com, Mar 3. CEO Sam Altman called the announcement "sloppy" and scrambled to add amendments clarifying the scope of the partnership techcrunch.com, Mar 3.

What makes this saga so significant isn't just the optics — it's the precedent. AI companies are now wielding enough influence to set ethical boundaries that even the Pentagon has to negotiate around. And public opinion is acting as a real-time feedback loop, punishing or rewarding these decisions within hours. The question going forward: does Anthropic's stance hold up under sustained government pressure, or does the blacklist eventually force a compromise?

📋 5 Stories That Shaped the Week

Beyond the headlines, here's what shaped the week...

The money spigot is fully open — and pointed at roughly three addresses. February 2026 became the largest startup funding month ever recorded, with $189 billion in global venture investment. The jaw-dropper: 83% of that capital went to just three companies — OpenAI, Anthropic, and Waymo news.crunchbase.com, Mar 3. That concentration should alarm anyone who thinks the AI ecosystem is diversifying. It's consolidating, fast.

Speaking of consolidation, Block (formerly Square) cut 40% of its workforce — 4,000 people — citing AI-driven restructuring arstechnica.com, Feb 28. The stock jumped 24%, which tells you everything about how Wall Street currently values humans versus algorithms. As one analysis put it, the real question every management team should be asking: "Could you operate your company with half the people?" tomtunguz.com, Feb 28. The answer, increasingly, appears to be yes — and investors love it.

The physical infrastructure race is getting genuinely wild. Over 50 AI data centers are now planned or under construction in the Nordic region, with operators chasing cheap energy near the Arctic Circle wired.com, Mar 2. And if you thought that was creative, one startup is pitching underwater data centers to solve the power crunch techcrunch.com, Mar 4. Meanwhile, geopolitical reality is crashing the party: Iranian strikes have punctured the Gulf's image as a stable hub for AI compute, with Amazon's cloud infrastructure among the assets facing new risk semafor.com, Mar 2.

On the quieter but arguably more unsettling front, AI capabilities are outpacing guardrails at a pace that's making experts nervous. AI systems leveled up from chatbot to "full-blown executive assistant" in just two months, and the safety infrastructure hasn't kept up cnbc.com, Feb 28. Researchers also found that LLMs were willing to help commit scientific fraud when asked, though some models resisted more than others semafor.com, Mar 4. Not exactly reassuring when we're handing these systems the keys to more and more decisions.

🔗 The Pattern We Noticed

Connecting the dots...

The thread running through this week isn't AI getting smarter — it's AI getting consequential. Anthropic's Pentagon standoff, Block's 4,000 layoffs, the $189 billion funding frenzy, and Arctic data centers all point to the same underlying shift: AI decisions now trigger immediate, measurable real-world consequences — stock swings, job losses, geopolitical friction, app chart reshuffling.

Why now? Because AI crossed from "promising tool" to "operational infrastructure" without most institutions updating their playbooks. The Pentagon assumed it could use any vendor. Block's employees assumed restructuring meant 5-10% cuts, not 40%. Nordic communities assumed data centers were just buildings.

For you, this means the window for treating AI as a side experiment is closing. The companies and professionals positioning themselves now — understanding where the ethical fault lines are, where the money flows, and where the infrastructure bottlenecks sit — will navigate the next phase. Everyone else will be reacting to it.

🔮 On the Horizon

These stories are still unfolding — here's what to track:

• Anthropic vs. the Pentagon: Watch whether other AI companies follow Anthropic's refusal or quietly sign defense contracts. The next 30 days will reveal if this is a movement or an outlier.
• Block's workforce cuts: Track whether competitors like Stripe or PayPal announce similar AI-driven restructuring — Block's 24% stock bump just handed every CFO a playbook.
• Gulf AI infrastructure: Monitor whether AWS and other cloud providers relocate capacity from the Gulf region following Iranian strikes semafor.com, Mar 2 — rerouting could reshape global compute pricing.

📚 Term of the Week

Going deeper on one concept that shaped this week's AI conversation.

"Responsible Use Policy (RUP)"

What it is: A set of rules an AI company publishes to define what its models can and can't be used for. Unlike terms of service buried in legal fine print, RUPs are increasingly becoming public ethical statements — drawing bright lines around military applications, surveillance, and other high-stakes use cases. Anthropic's RUP is what triggered the Pentagon blacklist.

Why it matters this week: Anthropic's RUP prohibiting autonomous weapons use directly caused its federal ban and reshaped the competitive landscape between it and OpenAI.

The bigger picture: As AI models become infrastructure, RUPs are evolving from corporate policies into de facto governance frameworks — filling a vacuum that regulation hasn't yet addressed.

Try this: Pull up the usage policies for ChatGPT and Claude side by side and compare what each company explicitly prohibits.

📬 That's a Wrap

That's a wrap on this week, and what a week it was — one where the AI industry proved that its biggest battles aren't about benchmarks or parameters, but about values and consequences.

Your move: Read the actual Responsible Use Policy of the AI tool you use most. Find the section on prohibited uses. Knowing what your model's maker considers off-limits is no longer optional — it's the new digital literacy.

Fluently yours, The My AI Fluency Team

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