Vittorio Banfi

Are we overestimating the hype or underestimating the payoff? Amazon's founder, Jeff Bezos just called the AI boom an “industrial bubble”  But he says it could still save the world. At this year’s Italian Tech Week in Turin, Bezos dropped a paradox: “This is a kind of industrial bubble… but that doesn’t mean anything that’s happening isn’t real.  AI is real, and it is going to change every industry.” Here’s why that matters, and what we, as builders, marketers, and founders, can take away. 𝐓𝐡𝐞 “𝐁𝐮𝐛𝐛𝐥𝐞” 𝐈𝐬𝐧’𝐭 𝐁𝐚𝐝, 𝐈𝐭’𝐬 𝐏𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐯𝐞 - Valuations are soaring, capital is flowing, and fundamentals are often ignored. Bezos calls it an industrial bubble, and he means it as progress, not panic. - Good and bad ideas both get funded. - Infrastructure is built. - Talent floods in. - When the hype fades, the useful innovations remain. “When the dust settles… societies benefit from those inventions.” - Jeff Bezos 𝐓𝐡𝐞 𝐂𝐨𝐫𝐞 𝐓𝐞𝐜𝐡𝐧𝐨𝐥𝐨𝐠𝐲 𝐈𝐬 𝐑𝐞𝐚𝐥 - Bezos isn’t dismissing AI as hype. His point is the opposite. “AI is real, and it is going to change every industry.” The debate is no longer about if AI matters, it’s about how fast and how deep the transformation will go. 𝐁𝐮𝐢𝐥𝐝 𝐟𝐨𝐫 𝐒𝐮𝐛𝐬𝐭𝐚𝐧𝐜𝐞, 𝐍𝐨𝐭 𝐇𝐲𝐩𝐞 - Every boom tempts founders to chase momentum. Bezos cautions against it. - Don’t build for attention. - Build systems that last. - Focus on solving real problems, not showcasing prototypes. For founders and go-to-market leaders: prioritize business value, adoption, and measurable impact. When the market corrects, only the useful products endure. 𝐓𝐡𝐞 𝐎𝐩𝐩𝐨𝐫𝐭𝐮𝐧𝐢𝐭𝐲 𝐈𝐬 𝐆𝐢𝐠𝐚𝐧𝐭𝐢𝐜 - Bezos sees the long-term gain. “The benefits to society from AI will be gigantic.” Like the dot-com and biotech booms before it, much will be overhyped and short-lived. But the infrastructure and innovation that remain will redefine industries, and societies. ✅ 𝐓𝐚𝐤𝐞𝐚𝐰𝐚𝐲𝐬 𝐟𝐨𝐫 𝐘𝐨𝐮 & 𝐘𝐨𝐮𝐫 𝐓𝐞𝐚𝐦 - Ask: Is my AI play durable or just flashy? - Build for resilience, not just speed. - Fuel ambition with discipline, because even good bubbles need good fundamentals - Remember: when the hype subsides, the winners won’t be the loudest; they’ll be the most useful. Question for you: When the AI wave settles and the dust clears, which companies will still stand, those chasing headlines or those solving real problems?

129

91 commenti

After years of building #Koin, I still see the same pain points repeat: 🚨 ANRs 🐢 Slow startups 💥 Memory leaks These are not “nice to fix” issues! They define the success of your app. Kotzilla (kotzilla.io) is built on that observation: 🔹 Detect issues early 🔹 Turn runtime data into actionable insights 🔹 And now… use AI to suggest precise fixes, powered by your app’s own context Your #Kotlin app is now on X-ray, in Debug & Production. Thanks to your DI powered by Kotzilla, "Firebase Performance & Crashlytics combined — on steroids".

59

1 commento

AI just delivered a wake-up call to the $350B news industry. What happened at an Italian newspaper will transform journalism forever. Il Foglio took a $250k gamble on AI journalism - writing 4 pages of daily content with attitude and wit. The results shocked everyone. The AI exposed political hypocrisy, tackled complex economic reforms, and engaged with readers through witty responses. This forced every major publisher to respond. Bloomberg launched AI financial insights while AP now automates 3,000 reports monthly. The Independent committed $10M to AI initiatives. Meanwhile, traditional media bleeds: 2,700 journalism jobs vanished last year. 2.5 local papers close weekly. Ad revenue plummeted 67%. By 2025, expect news that adapts to your expertise level and explains topics in your industry's language. It will connect patterns across stories, translate instantly into 95 languages, and create personalized audio for your commute. But there's a critical detail everyone missed. AI can't read body language in interviews, build trust with sources, or sense tension in a room. It can't replace human judgment. The winners won't replace journalists. They'll augment human capabilities. After investing $50M in AI infrastructure, I uncovered the real challenge: Modern AI requires massive computing power - 10,000+ GPU clusters, petabytes of training data, and military-grade orchestration. The problem is stark: 40% of training runs crash, millions burn in wasted compute, and progress stalls for weeks. We solved this with technology that auto-recovers from crashes, deploys spare GPUs instantly, and maintains 24/7 operations. Our clients now see 60% lower training costs, 85% less downtime, and 3x faster deployment. The future isn't about replacing humans. It's about building AI that actually works. Follow for more insights on AI infrastructure and the future of media. Like and share if you learned something new.

30

Here's what I've learned building digital products at Fcode Labs and others that actually keep users coming back. The first 5 minutes of your app experience is very important. When was the last time you gave an app a second chance after a confusing first experience? This is what actually works: • Make onboarding interactive, not just a slideshow. Let users learn by doing, not just reading. • Personalize the experience. A tech-savvy developer needs different guidance than someone who's less comfortable with technology. The most overlooked part is the Analytics. Specially when you are in early stage. You need solid analytics to see exactly where users drop off. Are they stuck on the signup? Confused by the dashboard? Without data, you're just guessing. Recently, we saw a reduction of 40% users dropping in Linkify just by redesigning onboarding flow based on actual user behavior patterns. Onboarding isn't just about teaching features. It's about showing value quickly. Users need to think "This solves my problem" within minutes, not days. What apps or platforms do you think have the best first-time experience? #ProductDevelopment #UserExperience #Analytics

82

Apple just changed the rules for Web2App in the EU… again. Apple’s new DMA update looks like more freedom, but comes with layered complexity and hidden fees. Yes, you can now use Stripe, webviews, and external links inside your app. But only if you remove IAP and accept a new fee stack: → 2% Initial Acquisition Fee → 5–13% App Store Services Fee → 5% Core Technology Commission Net cost: 10–20% per transaction before payment provider fees. You can cut some of those fees, but only if you give up App Store services entirely (analytics, reviews, updates, discovery, etc). It’s no longer about compliance. It’s about choosing your tradeoff: higher cost, more complexity, or less support. These changes don’t make the App Store experience better, they make it harder to operate. And that’s exactly why Web2App is becoming more relevant than ever. At FunnelFox, we’re building the most complete Web2App product on the market. Not just for funnel creation, but for the entire web payments stack. #web2app #dma #appstore #payments #subscriptions #growth #funnelfox

29

1 commento