Brief

Demis Hassabis, CEO of Google DeepMind, discussed recent AI advancements in a Two Minute Papers video. He highlighted breakthroughs in areas such as protein folding and climate modeling, emphasizing the potential for AI to solve complex global challenges. The discussion also touched upon the ethical considerations and future directions of AI research. AI

Google Cloud has issued a warning about the increasing sophistication of cyberattacks that leverage AI. The company is also expanding its "Google AI Pro Introduction Program" to accommodate up to ten participants. These initiatives highlight Google's focus on both the security implications of AI and the promotion of its AI technologies. AI

Demis Hassabis, co-founder of Google DeepMind, believes humanity is approaching the singularity, a point of rapid technological advancement. In contrast, Yann LeCun, a prominent AI researcher, argues that current AI systems do not exhibit true intelligence. Oriol Vinyals, co-lead of Google's Gemini, offers a middle ground, suggesting that while today's models are advanced, they still lack crucial learning and breakthrough capabilities. AI

Kunlun Wanwei Chairman and CEO Fang Han stated that experience is no longer a competitive advantage in the AI era, as AI can easily replace roles that are closed-loop and fault-tolerant. He believes that jobs requiring high judgment and accountability, such as those involving storytelling, idea generation, defining beauty, system building, and paradigm shifting, will remain irreplaceable. Fang Han also noted that token consumption is becoming a new metric for AI strength, with significant disparities between average users and heavy users, and advised companies to aim for the second position in AI implementation to balance innovation costs and market capture. AI

Anthropic's Code with Claude is demonstrating a future where developers increasingly rely on AI for coding tasks, with many admitting to shipping AI-generated code without thorough review. Google's recent I/O event highlighted a shift in AI-driven science, moving towards agentic, LLM-based systems like Gemini for Science, potentially reducing reliance on specialized tools. This trend is also reflected in the growing momentum behind 'world models' in AI research, aiming to imbue systems with a deeper understanding of the physical environment. AI

A former researcher from Google DeepMind has cautioned that relying solely on benchmarks may not be sufficient for advancing AI safety. The expert suggests that current evaluation methods might not adequately capture the complex risks associated with increasingly capable AI systems. This perspective highlights a potential gap between performance metrics and the actual safety of AI development. AI

A former Google DeepMind researcher has cautioned that relying solely on benchmarks is insufficient for ensuring the safety of advanced AI systems. The researcher emphasized that benchmark performance does not directly translate to real-world safety or true general intelligence. This perspective highlights the need for more comprehensive and robust evaluation methodologies beyond current standardized tests. AI

Google DeepMind CEO Demis Hassabis is exploring the concept of "virtual cells" as a next frontier, aiming to leverage AGI for fundamental breakthroughs in human biology and drug discovery. Separately, reports indicate that Russian-speaking actors are exploiting a jailbroken version of Google's Gemini AI to facilitate cryptocurrency wallet hacks, with at least one MAGA supporter falling victim. AI

Michal Valko, a computer scientist with over 30 years of experience in artificial intelligence, has worked with major tech companies like Intel, Meta, and Google DeepMind. He specializes in designing autonomous algorithms for deep reinforcement learning and self-supervised learning, aiming to create energy-efficient modular models and automate scientific discovery. Valko believes AI will become a common skill rather than a competitive advantage, prompting a re-evaluation of human value in an AI-driven world. AI

A hacker group known as TeamPCP is reportedly conducting large-scale attacks on open-source code repositories, with GitHub being a recent target. This group is allegedly poisoning software supply chains through these attacks. Separately, Google DeepMind CEO Demis Hassabis spoke at Google I/O, suggesting that the field of AI-driven science is rapidly advancing and is on the cusp of a significant transformation. AI

Researchers are suing the Trump administration over a visa policy that they claim restricts foreign-born workers studying online safety and content moderation. Separately, climate tech companies are shifting focus to critical minerals, with Boston Metal raising $75 million for this new strategy amid waning support for industrial decarbonization. The field of AI is seeing advancements in "world models," systems designed to understand the physical environment, with contributions from Google DeepMind, Fei-Fei Li's World Labs, and Yann LeCun's startup. AI

An online event on June 18th will explore how Anthropic's Claude AI can revolutionize e-commerce through live demonstrations of its code and cowork capabilities. Separately, Google is offering YouTube Premium Lite for free to users who subscribe to its AI plan, which costs 2900 yen per month. AI

Demis Hassabis, CEO of Google DeepMind, believes that widespread AI-driven job cuts are a misguided approach. He argues that AI should be used to augment human capabilities and create new opportunities rather than simply replace existing roles. Hassabis emphasized that the focus should be on leveraging AI to boost productivity and innovation, ultimately leading to economic growth and new job creation. AI

Google's I/O 2026 keynote featured a conversational Gmail bot and discussions from DeepMind leadership about the future of AI. Separately, the "Test of Time" update for Civilization 7 has been released by Firaxis, introducing a significant suite of changes to the game. AI

The Cannes Film Festival became a battleground for the film industry's views on artificial intelligence, with directors like Darren Aronofsky advocating for its use as a creative tool. Aronofsky, whose studio Primordial Soup is using AI in projects with Google DeepMind, argued that AI is additive and can solve production challenges, comparing its integration to the advent of sound or visual effects. Conversely, filmmakers like Guillermo del Toro expressed strong opposition, stating they would AI

A recent controversy surrounding Anthropic's Claude model, dubbed "Claude Mythos," has sparked discussions about the necessity of mindset shifts and accelerated countermeasures in AI development. Concurrently, Demis Hassabis of Google DeepMind has voiced his opposition to the notion that AI will render human developers obsolete, emphasizing the continued importance of human expertise in the evolving AI landscape. AI

Demis Hassabis, CEO of Google DeepMind, stated that solving mathematical problems like those posed by Paul Erdős does not equate to true invention. He believes that genuine invention involves creating novel concepts and technologies, rather than just finding solutions within existing frameworks. Hassabis emphasized that while mathematical prowess is valuable, it's distinct from the kind of groundbreaking innovation that drives technological progress. AI