Unsere AI Predictions für 2026 – Siri, IPOs & World Models | AI Talk by Clemens & Jakob

Unsere AI Predictions für 2026 – Siri, IPOs & World Models | AI Talk by Clemens & Jakob

🎙 Clemens Wasner & Jakob Steinschaden 👥 242 📅 January 6, 2026 ⏱ 60 min 👁 70 📄 expert opinion 🧭 2026-08-16
Available in: English (current) Français

Keywords

SiriGeminiAnthropic IPOPerplexityOpenAI advertisingChinese AIbenchmarksmedical AIworld modelsrobotics unicornventure capital

Summary

In this episode of AI Talk, hosts Clemens Wasner and Jakob Steinschaden present their 11 predictions for AI in 2026. They start with Clemens’ bold claim that Siri will finally become good thanks to a partnership with Google’s Gemini 3, integrated in iOS 26.4. Jakob predicts real-time video personalization, where content adapts to the user in real time. Clemens expects an Anthropic IPO, arguing it is more stable than OpenAI. Jakob foresees Perplexity being acquired by a telecom company like Deutsche Telekom or SK Telecom. They both agree OpenAI will introduce advertising, causing a backlash but not affecting user numbers. Jakob bets a Chinese model will top the LMSYS Chatbot Arena, while Clemens doubts it. They discuss the decline of traditional benchmarks, a potential medical breakthrough, the rise of world models and continuous learning, the first European AI robotics unicorn, and the transformation of venture capital due to AI. The episode ends with a sneak peek at the future of the AI Talk podcast, including a planned rebranding.

168 words

Critical Evaluation

Value of the Information & Strength of the Argument

The value of the information is moderate; it provides informed opinions from industry insiders but lacks empirical evidence. The argumentation is based on personal insights and observations of market trends, which are plausible but speculative. The hosts engage in a balanced discussion, considering counterarguments, which adds credibility. However, the predictions are subjective and not backed by data, limiting their scientific value.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor is low as the content is speculative. The hosts reference real companies and products (e.g., Gemini, Anthropic, Perplexity) but do not cite specific sources or studies. The title accurately reflects the content, focusing on predictions. The discussion is well-structured and coherent, but the lack of verifiable sources reduces its reliability.

129 words

Title / Content Match

The title accurately reflects the content: a discussion of AI predictions for 2026, including Siri, IPOs, and world models.

Quality & Reliability

6/10

The hosts are experienced AI entrepreneurs and commentators, but the content is speculative predictions without empirical evidence. They reference real companies and trends, but the claims are subjective forecasts.

Key Moments

Cited Sources

  • LMSYS Chatbot Arena — Referenced when discussing the ranking of AI models.
  • Synthesia — Mentioned as a startup working on video personalization.
  • Anthropic — Discussed in the context of a potential IPO.
  • Perplexity — Discussed as a potential acquisition target.
  • OpenAI — Discussed regarding advertising and user growth.

Concurring Sources

Contribution & Novelties

The episode offers a curated set of predictions for AI in 2026, providing a snapshot of industry expectations. It combines insights on consumer AI, market dynamics, and technical trends. The discussion on world models and continuous learning highlights emerging research directions.

Pour aller plus loin :

  • World Models — Background on the concept of world models in AI.
  • Continuous Learning — Overview of continuous learning paradigms.
  • LMSYS Chatbot Arena — The leaderboard referenced in the episode.
  • Anthropic — Company page for context on the IPO discussion.
  • Perplexity AI — Company page for context on the acquisition discussion.

97 words

Radar Profile

The radar profile shows moderate scores across all dimensions, indicating a balanced but not exceptional content. The highest score is in quantity of information, reflecting the number of predictions discussed, while technical depth and reliability are moderate due to the speculative nature.

Reliability 5/10