Anthropic rejects Pentagon's AI deal over surveillance concerns

As power users migrate to Claude.ai, How to Use It More Productively Anthropic's Claude has experienced a dramatic rise in early 2026, fueled by a high-profile Pentagon dispute, a viral Super Bowl ad campaign, and a wave of new model releases. Meanwhile, ChatGPT's market share has eroded as users migrate to Claude for its writing quality, ad-free commitment, and powerful productivity features. This report covers the key recent events and compiles a comprehensive list of tips—from prompting fundamentals to advanced agentic workflows—for getting the most out of Claude....

“No amount of intimidation or punishment from the Department of War will change our position on mass domestic surveillance or fully autonomous weapons.” Despite huge commercial incentives and the public spotlight, the AI major Anthropic is holding ground against the US govt. Brilliant and brave stuff. Link in the comments. It’s one of the most important developments happening in AI right now. Because *company values* and *digital trust* are going to be the fault lines of the AI revolution. Not the numbers of tokens, or front line models, or wizzy videos, or handwringing about the displacement of jobs. We MUST pay close attention to what we mean by ‘digital trust’ Because there are two types, and we mix them up all the time. Heres an example: “I trust my bank” -> Do they make your money available when you need it, and enable safe payments? Yes. -> But do they ‘accidentally’ forgot to tell you about going overdrawn, then charge you a fee? Also yes. Here’s the difference: 1. EXPERTISE TRUST - Can you do the thing well, reliably, scalably? - Will you do what you say you do? - Do you have the capabilities to make it work? 2. MOTIVE TRUST This is the important, and so often missing one: - Are your incentives aligned with mine? - Is your reputation consistent? - Do you weaken safety, security, privacy or customer/ partner/ citizen protection for your own benefit? Expertise vs motives. Look at the US govt example again. Fun fact: Once Anthropic was shown the door by the US Dept of War, ChatGPT swept in right after and sealed a commercial deal in its place. In spite of the lack of guardrails around safety, and the absence of the principles around autonomous killing and surveillance. As they say, once someone shows you who they are, believe them. I think we now have all the evidence we need about values and digital trust for both Anthropic and ChatGPT. So place your bets about who is going to win long term. More in this week’s Customer Futures newsletter. Link below👇

Thoughtful post from Jamie Smith💡. Totally agree"*company values* and *digital trust* are going to be the fault lines of the AI revolution." This point was highlighted in my dissertation (https://lnkd.in/gHqwgK3p all segments, trust was a defining factor that would lead people to adopt emerging sovereign identity technologies. I think we can reliable exend this to pretty much every engagement. This also reminds me of Don Peppers's book "Extreme Trust" (2016) where he argued that trust would be the key differentiator between businesses.

Fraudsters have always been creative — but now they’ve got AI scaling their operations. In the past two years, AI generated identities have increased 17%, enabling synthetic profiles that mirror legitimate customers across credit, device and behavioural signals. They blend authentic data with fabricated elements so effectively that traditional controls struggle to distinguish the difference. Spot the imposters before they embed themselves in your customer base: http://transu.co/6042hAEoY

📣 The Hiya Inc. 2026 State of the Call report is live 🎬 Key takeaways: 86% of consumers do not answer unknown calls!!! 🫨 31% of consumers say they received a deepfake voice call in the past year 📲 Fraud is scaling. Synthetic voice is advancing. Trust decisions are happening in milliseconds For businesses, that means a branded and verified identity is no longer optional if they want customers to pick up. For mobile operators, protection and trust at the network level are now competitive differentiators. Trust has to be built in before the phone rings. If answer rates and subscriber trust matter to you, this global report is worth a read. Download it here: https://hubs.ly/Q045bFyp0 #StateOfTheCall #VoiceSecurity #Telecom #AI #FraudPrevention

 An AI Company Said No to the Pentagon - and Got Blacklisted What happened: On Friday, February 27, Anthropic refused to drop two red lines when negotiating a Pentagon contract: no domestic mass surveillance and no autonomous weapons without a human making the final call. Hours later, Defense Secretary Hegseth designated Anthropic a "supply-chain risk to national security" - language historically reserved for foreign adversaries like Huawei, never a US company. Trump then ordered all federal agencies to stop using Claude within six months. Anthropic announced it would challenge the designation in court, calling it "legally unsound and a dangerous precedent." Dan's take: Here's what this means beyond the headline: people didn't just feel bad for Anthropic — they went to the App Store. Claude hit #1 in the US free chart over the weekend, and verified ChatGPT subscribers posted cancellation threads on Reddit. For the first time, consumers are picking AI tools based on values, not features. That's a new variable in this industry, and nobody has fully priced it in yet. The test: Claude at #1 on the App Store is a real data point. Whether those downloads become loyal users is the test I'll have a verdict on next week. Advertising on your profile: https://lnkd.in/dde_ZUzg

ChatGPT just handed your data to the Pentagon. And people are FURIOUS. In case you missed it, OpenAI quietly signed a deal with the U.S. Department of Defense on a Friday, giving the Pentagon access to deploy ChatGPT inside classified military networks. The public found out. And lost it. 295% spike in uninstalls. Overnight. 775% surge in one-star reviews. In a single day. 2.5 MILLION reported uninstallations. And counting. Think about that for a second. Every prompt you've ever typed. Every question you've ever asked. Every private thought you fed into that chatbox. This deal just changed who has access to all of it. Meanwhile, Anthropic, the company behind Claude, looked the government in the eye and said NO. No mass surveillance. No autonomous weapons. Not until AI is actually ready for it. Within 48 hours, Claude downloads surged 37%. Then 51%. Not because of a campaign. Because people were paying attention. Do you think OpenAI crossed a line? Or is this just the cost of doing business in the AI era? Drop your take in the comments and share this message with others. Your network deserves to know.

Anthropic got blacklisted by the Pentagon. Days later, Claude was the most downloaded app in America. The government called Anthropic a "supply-chain risk to national security." Because Anthropic refused to remove AI safety guardrails for military use. Most companies would have caved. They didn't. They filed a lawsuit instead. 149,000 daily downloads. vs ChatGPT's 124,000. First time Claude has ever topped ChatGPT. Not by a little by 20%. Daily active users tripled since January. Paid subscribers doubled. The Pentagon ban was the best marketing Anthropic never paid for. When you stand for something publicly even when it costs you a government contract people notice. The general public saw a company choose principles over money. And they responded. Most AI companies talk about safety. Anthropic walked away from a military deal to prove it. The AI race isn't just about whose model is smarter. It's about whose values you trust with your data, your work, your future. Claude just won that conversation. At least for now. Have you switched to Claude yet or are you still on ChatGPT? #vrajdattani #artificialintelligence #anthropic #claude

Did you know that 93% of senior executives admit to using AI tools without IT oversight? While security teams are busy building perimeters around infrastructure, a massive "blind spot" is growing at the top. Executives bypass the very procedures they demand from employees to "move faster." But leaking strategic data through a free instance of ChatGPT is irreversible. Is your board unknowingly exposing the company to risk? I’ve built a free 30-second audit to diagnose your Shadow AI exposure 👇 Link in comments

Did you know that 93% of senior executives admit to using AI tools without IT oversight? While security teams are busy building perimeters around infrastructure, a massive "blind spot" is growing at the top. Executives bypass the very procedures they demand from employees to "move faster." But leaking strategic data through a free instance of ChatGPT is irreversible. Is your board unknowingly exposing the company to risk? I’ve built a free 30-second audit to diagnose your Shadow AI exposure 👇 Link in comments