Deepmind CEO Hassabis: World models are the future, but the AI bubble is real

OpenAI has introduced GPT‑5.1-Codex-Max, a new coding model available in its Codex developer environment. This model enhances AI-assisted software engineering with improved reasoning, efficiency, and interactive capabilities. It replaces GPT‑5.1-Codex as the default model, designed to handle complex software development tasks across multiple contexts.

Google's latest AI model, Gemini 3, reportedly surpasses competitors in various benchmark tests. However, concerns about its reliability persist, raising questions about the sustainability of the current AI hype.

Meta has secured a significant legal victory against the US Federal Trade Commission (FTC) in an antitrust case, which could allow the company to resume acquiring startups after a period of regulatory caution. This ruling may also impact other major tech firms like Google and Microsoft, who have been hesitant to pursue acquisitions due to increased scrutiny under FTC Chair Lina Khan.

Google has launched Gemini 3, its most advanced AI model to date, which promises enhanced reasoning capabilities, 'vibe coding', and the ability to perform complex multi-step tasks. This development raises questions about the potential impact of AI on human jobs and the future of work.

Demis Hassabis, CEO of Google DeepMind, discussed the advancements of Gemini 3, Google's latest AI model, in an exclusive interview. The model is positioned as a significant leap in AI technology, promising enhanced reasoning and coding capabilities. This development marks a pivotal moment as Google enters a new era of intelligence with Gemini 3.

TikTok is set to empower users by allowing them to reduce the amount of artificial intelligence-generated content in their feeds. The platform, which currently hosts over 1 billion AI videos, is testing this feature over the coming weeks before a global rollout. This initiative comes in response to the increasing prevalence of AI-generated content, driven by new tools like OpenAI's Sora and Google's Veo 3.

AfriSpeech-MultiBench is introduced as the first domain-specific evaluation suite designed for over 100 African English accents across more than 10 countries. This benchmark suite spans seven application domains, including Finance, Legal, Medical, General dialogue, Call Center, Named Entities, and Hallucination Robustness. It aims to address the lack of publicly available application-specific model evaluations that consider Africa's linguistic diversity. The suite benchmarks various ASR and LLM-based speech recognition systems using both spontaneous and non-spontaneous speech from open African…

This study evaluates the effectiveness of various large language models (LLMs) in restoring diacritics in Romanian texts, a crucial task for text processing in languages with rich diacritical marks. The models tested include OpenAI's GPT-3.5, GPT-4, Google's Gemini 1.0 Pro, and Meta's Llama family, among others. Results indicate that GPT-4o achieves high accuracy in diacritic restoration, outperforming a neutral baseline, while other models show variability. The findings emphasize the importance of model architecture, training data, and prompt design in enhancing natural language processing to…