I’ve seen it happen before — not with AI, but with every shiny new tech that sweeps Silicon Valley. The pattern starts the same way: a wave of excitement, a flood of half-baked tools, and then a rush of companies scrambling to slap the label on their product without asking if it’s actually useful. This time it’s AI’s turn, and it’s worse than most because the tech is both overhyped and legitimately powerful. The result? Companies are developing what I call AI psychosis — a condition where they can’t tell the difference between a real opportunity and a marketing gimmick anymore.

What does AI psychosis look like in real companies?

Take a look at the last ten startups that pitched me in the past three months. Every single one had an AI angle, even when the problem they solved had nothing to do with artificial intelligence. One sold AI-powered invoicing software to accountants. Another built an AI chatbot that answered customer service calls for a local car dealership. Neither product needed AI to work, but both teams insisted it made their service “smarter” or “more cutting-edge.” The invoicing tool took three extra months to build and cost 40% more because of the AI wrapper. The chatbot got so confused by simple requests that the dealership switched it off after two weeks and hired a human instead.

I’ve talked to engineers and product managers at firms big and small who admit they’re adding AI features just because their investors expect it. One friend at a mid-sized SaaS company told me they spent six weeks building an AI summary tool for customer feedback — only to discover their own support team never used it. The tool was slower than a human reading the emails, but the CEO insisted it “looked good on paper.” Another friend at a logistics startup said they built an AI demand-forecasting model that cost $200,000 a year to run. Their actual sales team could predict demand just as well by looking at spreadsheets. The model was shut down after a quarter.

Why are smart people falling for it?

The psychology is simple: fear of missing out mixed with pressure to innovate. Every tech conference this year had at least one session titled “How to integrate AI into your business” or “Future-proof your company with AI.” The message is clear: if you’re not using AI, you’re falling behind. Add to that the fact that AI startups raised $56 billion globally in the first half of 2024 alone, according to PitchBook, and you’ve got a perfect storm. Companies don’t want to be left out, so they throw AI at problems whether it fits or not.

The problem gets worse when you factor in how hard it is to admit you don’t understand something. Silicon Valley culture glorifies confidence above all else. No one wants to say, “We don’t actually know if this AI feature helps anyone.” So instead, they double down on the hype. Even well-intentioned teams start using AI jargon so much that they lose track of what they’re actually trying to achieve. One founder I know kept saying his team was “training the model” when they were really just tweaking a rule-based chatbot. He believed it so much that he raised $8 million on the promise of an “AI-native” product.

The cost isn’t just money — it’s time and trust

The real damage isn’t just the wasted engineering hours or the bloated budgets. It’s the erosion of trust inside companies. Teams that once collaborated well start arguing over whether something is “AI enough” to ship. Engineers feel pressure to add AI features that make the product worse. Product managers waste weeks writing user stories about AI behaviors that don’t exist yet. Customers get confused by half-baked AI integrations that break more than they help.

I’ve seen this before with blockchain in 2017 and Web3 in 2021. The difference this time is that AI actually works in some cases. That makes the delusion harder to spot. It’s not that AI is useless — it’s that it’s only useful for specific problems where the data is clean, the use case is clear, and the cost is justified. Most companies aren’t in that position. They’re treating AI like a magic wand, and that’s dangerous.

What happens next?

Eventually, the hype will crash into reality. Investors will start asking harder questions. Customers will stop paying for AI-washed products that don’t work. Teams will burn out from chasing trends instead of solving real problems. The companies that survive will be the ones that treat AI like a tool — not a religion. They’ll ask: Does this actually solve a real problem? Is the benefit worth the cost? Can our team maintain it without burning out?

The best tech isn’t the one that shouts AI the loudest. It’s the tech that quietly makes life easier for the people who use it. That’s a lesson some companies are about to learn the hard way.

What You Need to Know

  • Source: Hacker News
  • Published: May 15, 2026 at 20:26 UTC
  • Category: Technology
  • Topics: #hackernews · #programming · #tech · #ai-hype · #ai-psychosis · #ai-delusion-in-tech

Read the Full Story

This is a curated summary. For the complete article, original data, quotes and full analysis:

Read the full story on Hacker News →

All reporting rights belong to the respective author(s) at Hacker News. GlobalBR News summarizes publicly available content to help readers discover the most relevant global news.

Curated by GlobalBR News · May 15, 2026

🇧🇷 Resumo em Português

O mundo da tecnologia vive um frenesi sem precedentes com a inteligência artificial, mas especialistas alertam: muitas empresas estão mergulhando de cabeça no hype sem medir os riscos reais. Em meio a promessas mirabolantes de automação e eficiência, gigantes do setor parecem esquecer que a IA ainda engatinha em questões como viés, privacidade e transparência — problemas que, no Brasil, podem ter consequências ainda mais graves.

No Brasil, onde a regulação de dados é tema recente e a desigualdade digital ainda é uma realidade, a corrida desenfreada pelas soluções de IA levanta preocupações sérias. Empresas locais e multinacionais operando no país muitas vezes adotam modelos estrangeiros sem adaptação, ignorando que problemas como discriminação algorítmica ou vazamentos de dados afetam diretamente milhões de brasileiros, especialmente nas periferias. Além disso, a dependência de tecnologias estrangeiras pode reforçar uma lógica colonial digital, colocando em xeque a soberania tecnológica do país.

Enquanto isso, a pergunta que fica é: até quando o Brasil vai acompanhar essa onda sem questionar suas bases? A hora de discutir regulação, ética e limites parece não poder mais esperar.

🇪🇸 Resumen en Español

La industria tecnológica parece haber sucumbido al hechizo de la inteligencia artificial, con empresas de todos los tamaños abrazando con fervor soluciones que, en muchos casos, carecen de una aplicación práctica real. Expertos como el veterano periodista tecnológico Benedict Evans advierten que, en su afán por no quedarse atrás, muchas corporaciones están cayendo en la “delusión de la IA”, un fenómeno en el que la obsesión por la innovación supera al análisis crítico de su verdadero valor.

Este entusiasmo desmedido no es solo un capricho pasajero, sino un riesgo estratégico con consecuencias tangibles para el mercado hispanohablante. En un contexto donde la accesibilidad y la utilidad deben primar sobre el marketing, muchas empresas latinoamericanas y españolas podrían malgastar recursos en proyectos de IA sin retorno tangible, desde chatbots que no resuelven problemas reales hasta sistemas de automatización que generan más frustración que eficiencia. La delusión radica en creer que implementar IA es sinónimo de modernidad, cuando, en realidad, su éxito depende de una integración inteligente y centrada en las necesidades concretas de los usuarios.