MIT study: 95% of GenAI deployments lack value

💡 A lot of buzz in recent weeks about the MIT research that found that 95% of GenAI deployments aren’t yet delivering value. But are 95% of AI projects really worthless? Great article from my colleagues Catherine Brien, Tim Roberts, and George Cevora. It’s a useful reminder of what we’ve seen across industry: optimism is high, but execution is where most fall short. As we explored in our “Beyond the Hype” work with World Retail Congress, the real challenge isn’t whether AI has potential; it’s how to turn that potential into impact at scale. To succeed with AI, organisations need to stay disciplined and focused on achieving tangible results that drive real impact. https://lnkd.in/d6ifkqGb

Are 95% of AI projects really useless? A very timely article from my AlixPartners colleagues Catherine Brien, Tim Roberts, and George Cevora: https://lnkd.in/ettv9SxW A recent MIT study found that 95% of GenAI deployments aren’t yet delivering value. Whilst optimism is high, disciplined execution is where many are falling short. There's no doubt the potential is there; but businesses are struggling to turn potential into impact, and crucially at scale. How to overcome the challenges? A focus on building the right foundations, aligning technology to strategy, and leveraging AI to enhance - not substitute - ambition and impact

This is super interesting and a very useful take on AI right now. One particular truism that resonates for me is - “Organize for success: Centralized leadership, decentralized action” … it’s at the individual and team level where organisations really deliver benefits.

95% of enterprise AI projects fail to deliver ROI — not because of the models, but because of execution. The problem isn’t the tech — it’s execution. 🚫 Top killers of AI ROI: ⚡ No clear value hypothesis ⚡ No baselines or A/B tests ⚡ Pilots stuck in labs, not production ⚡ Adoption gap — users don’t trust or use it The fix? Treat AI like any other business initiative: tie it to measurable outcomes, prove impact with data, and embed it into workflows. AI success = governance + measurement + adoption → not just models. #AI #EnterpriseAI #ROI #DigitalTransformation https://lnkd.in/egkDbbun

95% of Companies see zero return on $30B+ GenAI spend This comes from MIT’s 2025 hard data. Most AI pilots fail to move beyond experiments. Why? Because deploying technology isn’t enough. The gap lies in integration: AI tools without memory, workflows left unchanged, and no clear link to strategic goals. I break down the full MIT report in my latest article: https://lnkd.in/dgzqXWJZ P.S. If you want to cut through the noise and accelerate AI impact, DM me to learn how my AIG-OS™️framework drives real AI deployment and governance.

Trust VectorMind enterprise #AI to fill the gap between AI promise and AI impact! In a recent study titled 'The GenAI Divide: State of AI in Business 2025' by MIT Professional Education (https://lnkd.in/g5espHsG ), findings were that 95% of organisations are seeing no measurable return on their AI projects, with only a rare 5% of pilots delivering millions in value. In other words, most enterprise AI initiatives never scale beyond flashy proofs-of-concept. The reasons are telling: ✅ brittle workflows, ✅ lack of contextual learning, and ✅ poor integration into day-to-day operations In this article, we investigate how CohesionX's success with VectorMind correlates with these conclusions. READ IT HERE ⤵️ https://lnkd.in/dHCrhdYh

”about 5% of AI pilot programs achieve rapid revenue acceleration; the vast majority stall, delivering little to no measurable impact on P&L.” Interesting - and disappointing - results from an MIT study; though aligns with my observations. https://lnkd.in/dqZNUdKM

MIT report: 95% of #generativeAI pilots at companies are failing "How companies adopt #AI is crucial. Purchasing AI tools from specialized vendors and building partnerships succeed about 67% of the time, while internal builds succeed only one-third as often." https://lnkd.in/gz9Cyvg3

MIT’s latest research: 95% of generative AI pilots are falling short, not because of the models, but because of a “learning gap.” When it comes to delivery, vendor partnerships succeed far more often than internal builds, especially in regulated sectors.   The takeaway? Success isn’t about running pilots, it’s about embedding AI where it matters and helping organizations adapt. That’s why our work starts with education first, so adoption drives real outcomes.