EmTech AI 2026 explored how artificial intelligence is evolving from standalone tools into integrated platforms that orchestrate workflows across industries. Experts discussed the implications of AI platforms becoming the central infrastructure for business operations, decision-making, and innovation.
Background
- MIT Technology Review’s EmTech conference is a major annual event where industry leaders, researchers, and policymakers discuss emerging technologies. The 2026 edition focused on AI platforms.
- An "AI platform" refers to a foundational system (like OpenAI’s ChatGPT, Google’s Gemini, or Meta’s Llama) that other developers build applications on top of, analogous to how iOS or Android are mobile platforms.
- The article signals a shift from standalone AI tools and chatbots toward a platform ecosystem: companies are competing to become the default infrastructure that powers everything from enterprise software to consumer apps.
- Key context: The AI landscape in 2026 is marked by fierce competition between tech giants (Microsoft, Google, Meta, Amazon) and startups, regulatory debates over safety and antitrust, and the question of whether AI platforms will be open-source or proprietary.
Zcash's formal verification efforts have eliminated undetectable counterfeiting bugs from the new Ironwood pool, ruling out such vulnerabilities up to the underlying cryptographic assumptions, as part of Project Tachyon.
Zcash testnet is updating for Ironwood with two independently developed consensus implementations, one by Valar Group (in audit) and another by the Zcash Foundation. Users can try a desktop wallet fork with migration code, and Keystone dev device users can update firmware to test signing 11+ transactions with a single QR code.
A tweet from Eric Migicovsky claims that the Apple Watch lacks a certain capability, contrasting it with the Pebble smartwatch, as illustrated in an attached image.
In this weekly update recorded from a riad in Marrakech, Troy Hunt discusses the futility of trying to delete personal data from the internet, comparing it to removing urine from a swimming pool, while covering recent data breaches.
Raymond Chen explains a specific type of Control Flow Guard (CFG) check that combines validation and function calling into a single operation, describing how this "combined validate and call" approach works as a two-in-one package within Windows security mechanisms.