🕶️ AI chips, chat-for-cash, and Siri’s great resignation

The “AI roll-up” model — startups buying struggling companies and boosting them with AI — is gaining ground among VCs like General Catalyst, Thrive, Khosla, and Bessemer.

On the Tech.eu podcast, founders Din Bisevac (Buena), Constantin Schröder (Arbio), and Dan Lifshits (Dwelly) discussed the trend. Schröder said property and rental management are ideal for AI because they involve coordinating owners, guests, and service providers.

Bisevac noted early integration challenges after acquisitions — from skeptical staff to resistant customers — but found they ease once the company delivers a better product and service. Read more!

What will your job look like in 2035? We explore three possible futures shaped by AI; one where work gets easier and better paid, one where nothing really changes, and one where things get worse. This video breaks down the data, the risks, and what needs to happen to make the best-case scenario a reality.

ByteDance’s chatbot Doubao just pulled ahead of DeepSeek to become China’s most-used AI app — 157 million monthly users compared to 143 million in August.
Its edge isn’t raw intelligence; it’s design. Doubao feels friendly, works across text, image, and video, and is tightly woven into Douyin (China’s TikTok). Users can turn chats into shareable clips, making the app part of the social loop. It even ranked fourth globally in a16z’s August AI index.

The takeaway: great UX beats great specs. When an AI lives inside a huge attention network and feels effortless to use, people stick around.

We’re heading toward a world where chat, creation, and agents blend into everything — social feeds, shopping apps, even cars. For startups, that means the real battles are onboarding, trust, and virality, not model power. Big platforms like YouTube, Instagram, and Tesla already have the networks to build all-in-one “everything apps.” Expect regulators to start worrying about how much data those ecosystems trap.