12 hours ago I posted: "I had Fable 5 for breakfast." 12 hours later, US politics ate Anthropic for dinner. Not because of a catastrophic failure. Not because of mass misuse. But because of what @AnthropicAI calls a "narrow potential jailbreak." That's the story. AI isn't competing only on intelligence anymore. It's competing on regulation, compliance, and geopolitics. The next frontier model won't just need better reasoning. It will need permission to exist. Welcome to the new AI race. #AI #Anthropic #Claude #AIRegulation

Paris. AI. Coffee. That's my agenda for #VivaTech 2026. I'll be around all week meeting founders, business leaders, creators, AI enthusiasts, technology partners, and anyone curious about the future of AI. Networking has always been one of my favorite parts of events like this. I'll be at my booth (3H24-002) on the last day, so if you're attending, come by, say hi, take a photo, and let's talk AI, Microsoft 365 Copilot (or any other AI), AI agents, automation, business growth, partnerships, or simply exchange ideas. If you're attending VivaTech, reply below or send me a DM. See you in Paris. AIDARIUS #VivaTech2026 #Networking

Let's have some fun they said, let's use some claude they said. Sounds familiar? p.s. the main idea is not mine :) #ArtificialIntelligence #AIAdoption

Hey @X algorithm 👋 I'm looking to #connect with leaders, builders, and practitioners who believe one thing: AI doesn't fail at the technical layer first. It fails at the decision layer. If you're working on: • AI strategy & governance • Microsoft 365 Copilot • @GoogleAI AI ecosystem • @NVIDIAAI infrastructure and ecosystem • Enterprise and SMB's AI adoption • AI agents & automation • AI security & compliance • Scaling AI across organizations Let's connect. Reply with one thing: What's the biggest AI decision your organization still hasn't made? I'll start: Most companies don't have an AI problem. They still haven't decided who is responsible for AI.

AI is dramatically overrated. Most companies still can't get consistent business value from it, and human expertise remains the bottleneck. Change my mind.

Every role in IT taught me something useful for the AI era. The body transformation though… probably the most accurate KPI of corporate growth.... Now suddenly everyone is into longevity, fasting, cold plunges, and biohacking. 😄 What did your evolution look like?

The price of growth. The real enterprise question is not: “Are the numbers bad?” It’s: “Are they bad enough to break trust?” SMB consumers tolerate outages. Enterprises tolerate risk only until workflows become unpredictable. AI adoption does not fail at the technical layer first. It fails at the trust layer.....and trust scales much slower than hype.

ChatGPT or Microsoft 365 Copilot - your choice? ChatGPT wins on creativity and flexibility. No doubts. Microsoft 365 Copilot wins on enterprise integration and governance security. Interesting part? Many employees still prefer and use ChatGPT. Enterprises often choose Microsoft 365 Copilot because AI adoption is no longer just about prompts, it’s about security, workflows, permissions, and ecosystem. My take: ChatGPT = better thinking partner. Microsoft 365 Copilot = better enterprise employee and assistant AI follows decisions

AI is not failing in companies. Ownership is. Everyone is using AI. No one wants to own the consequences. Who actually owns AI decisions where you work? 🟡 CEO 🟢 IT 🔴 Legal ⚫ No one Reply with your answer.

Your business profile portrait photo. When did you last update it? Mine? 2–3 years ago. Old way: studio (€100–€300) photographer (€150–€400) Now in 2026: AI → studio-level portraits in minutes for free or a few cents. So… is professional portrait photography dead? No. But “professional” has changed. If you don’t know how to do it, follow me and comment AI-PHOTOGRAPHY and I’ll send you details. AI follows decisions.

Some conferences talk AI. Some talk tools. This one hit different.“Lifelong Learning: The New Skills Economy” Not even about AI (yes, a bit, but not the main course 🙂) I took one idea: → keep learning. All of your life. That’s the edge. AI follows decisions.

My brain exploded last week. At Vilnius University, I saw a presentation of a digital twin of a human heart. That was enough. This is not ChatGPT. This is AI simulating reality. Helping doctors decide before they touch a patient. Less content. More consequences. AI follows decisions.

I bought Premium on X and got the blue badge. Feels valuable. Let’s be honest: it’s distribution, not intelligence, not organic growth. Exactly what’s happening in AI: Tools are getting better. Decisions are not. AI follows decisions.

In 3 hours I’ll be watching the @nvidia GTC 2026 keynote 2 hours about AI infrastructure. Everyone talks about AI tools. But real progress happens in: compute memory energy And the real bottleneck isn’t technical. It’s decisions. AI doesn’t lead decisions. AI follows decisions.

It’s not about who scraped what. It’s about who signed off on using it. Models don’t create exposure. Decisions do. #AIFollowDecisions #AIDARIUS

AI doesn’t fail because it’s powerful. It fails when we let it decide what it should only assist. Define what AI can: recommend execute never own If you don’t, governance is theatre. with love from AIDARIUS.com #AIFollowDecisions

If an AI decision causes damage, ask one question: Who owned the decision? If no one can answer, you’ve already found the failure.

If leadership can’t name who owns data, decisions, and risk AI won’t scale.

AI doesn’t fail at the technical layer first. It fails at the decision layer. When decisions aren’t defined or owned, open-source personal AI assistants like @openclaw stop being helpers and become security gaps. Models and AI agents aren’t the risk. Ungoverned decisions are.