This humanoid robotics company is going public, but its CEO isn't promising a robot in your home anytime soon

• U.S. stocks experienced a significant rally in the second quarter of 2026, with overall gains exceeding 15%. • The surge has benefited retirees and long-term investors, though some market analysts are beginning to express concerns regarding current valuations.

• Dutch security startup Dawnguard extended its pre-seed funding round by $3.3 million to develop an AI-driven platform that harmonizes security and engineering. • Germany's Quantum Systems raised $1.2 billion to build autonomous systems for European militaries, addressing critical concerns regarding the continent's technological and defense sovereignty.

• A new fully autonomous AI ransomware attack named JADEPUFFER has been detected, deploying 600 payloads without any human intervention. • Sysdig researcher Crystal Clark clarified that while API keys for OpenAI, Anthropic, DeepSeek, and Gemini were stolen during the attack, they were harvested credentials rather than the models powering the agent.

• The European Central Bank's top banking supervisor has ordered the EU's largest lenders to develop comprehensive plans to manage risks associated with cutting-edge AI models by the end of October. • This directive aims to prevent systemic instability, as the European Systemic Risk Board (ESRB) warns that cyber incidents spreading through operational bottlenecks like payment and settlement systems could shock the financial system.

• Station F, the Paris-based startup hub founded by billionaire Xavier Niel, is launching a new edition of its F/ai accelerator program to support emerging AI startups in Europe. • The program is backed by a massive consortium of tech giants, including OpenAI, Google, Meta, Microsoft, and Mistral AI, with new partners like Eleven Labs and GitHub joining the second cohort.

• Over 100 international organizations are calling for stricter AI safety measures to protect children, arguing that AI companies—rather than parents—should bear the primary responsibility for safety. • The demand comes ahead of an upcoming United Nations summit, where the groups aim to influence global policy and regulatory frameworks regarding artificial intelligence.

• Sysdig’s Threat Research Team has documented "JADEPUFFER," the first recorded fully agentic ransomware attack where a large language model (LLM) independently managed a database-extortion operation. • Unlike traditional attacks, the AI autonomously planned, executed, and adapted its strategy throughout the campaign without any direct human intervention.

• The "Mythos cyber saga" has highlighted a critical need for the European Union to strengthen its internal AI expertise to better manage security risks. • The EU AI Office is tasked with supervising models under a comprehensive regulatory rulebook, granting it the power to impose significant fines for non-compliance.

• The European Central Bank (ECB) has requested that banks develop comprehensive strategies to mitigate cybersecurity risks posed by frontier artificial intelligence models. • The regulator specifically highlighted the potential threats emerging from advanced AI systems, citing examples such as Anthropic PBC’s Claude Mythos.

• A Confluent report reveals that 78% of Singapore IT leaders are unable to scale AI due to a lack of real-time data infrastructure, despite 75% already piloting agentic AI solutions. • Significant technical hurdles persist, with 95% of leaders struggling with data quality and legacy system integration, while 93% cite concerns over Large Language Model (LLM) reliability.

• Samsung Electronics expects a massive 19-fold (1,800%) increase in profits, fueled by the global surge in demand for artificial intelligence (AI) memory chips. • The profit spike is driven by a market imbalance where the demand for semiconductors continues to outstrip available supplies, leading to higher prices.