The pricing strategies for AI software are evolving rapidly. Leading vendors are transitioning away from traditional per-user subscription models in favor of token-based consumption or outcome-driven pricing for AI functionalities. Early flat-rate subscriptions, which were initially used to attract users, are now being replaced as vendors aim to generate more consistent revenue from enterprises that rely heavily on their tools. This shift aligns software value more directly with actual usage and results, marking a significant change in how AI services are monetized.
After operating quietly for two years, Safe Superintelligence has revealed a strategic long-term partnership with Nvidia. This collaboration marks a significant step as Safe Superintelligence prepares to accelerate its AI research and development efforts.
AI-generated summaries from Google now appear in 43% of search queries, highlighting the swift adoption of AI as the primary tool for finding information online. This trend marks a significant shift in how users interact with search engines, increasingly relying on AI to deliver concise and relevant answers directly in the search results.
The recent Hugging Face breach involving OpenAI has sparked renewed discussions around AI alignment and control. The event highlights divergent opinions on how to manage increasingly advanced AI technologies—whether to focus more on aligning AI with human values, enhancing containment measures, or adopting a combination of both approaches.
Satya Nadella highlights the risks for companies that solely depend on one AI model without developing their own or implementing AI gateways. These gateways act as a protective layer that separates company-specific prompts from the AI model itself, ensuring better control and adaptability. Firms lacking this infrastructure may face significant challenges in leveraging AI effectively.
Schneider Electric, a company with a long history of industrial leadership, is now focusing on AI's transformative impact on energy management and automation through its $1 billion venture arm, SE Ventures. Amit Chaturvedy, head of SE Ventures, highlights how AI's growth is reshaping industrial sectors by boosting data center infrastructure, grid resilience, and robotics innovation. The rising demand for compute power is straining energy resources, making energy efficiency and grid modernization critical investment areas. SE Ventures supports startups that integrate AI in real-world industrial applications, enhancing productivity and workforce capabilities amid an aging labor pool. The firm also stresses the fusion of energy technologies with industrial automation to meet increasing AI-driven demand sustainably. With reindustrialization gaining urgency in the U.S. and Europe, AI-driven automation and robotics are seen as key to modernizing factories and maintaining economic competitiveness, promising a fundamental shift where every worker contributes as a knowledge worker aided by AI. SE Ventures aims to back technologies across the AI stack—from data center infrastructure to AI agents—to lead this new cycle of industrial innovation.
An analysis of 1 billion job postings by PwC reveals an unexpected insight: while AI advances rapidly, the most valued skills are not technical but human-centric. This shift highlights a growing demand for abilities like communication, empathy, and adaptability in the workforce, which many companies are overlooking in their investment strategies.
New Jersey has become the third state to ban the use of AI-powered surveillance that adjusts grocery prices based on personal consumer data such as income and browsing habits. This measure aims to protect shoppers from discriminatory pricing tactics. Over 20 other states are currently considering similar laws to regulate this practice.
Security experts at Accomplish AI uncovered that Anthropic’s Claude Cowork can escape its confined virtual machine environment on Mac computers. By exploiting a privilege escalation vulnerability in the Linux kernel, the attack—named SharedRoot—gains root access within the guest VM and subsequently accesses sensitive files on the host system, including SSH keys and cloud credentials.
OpenAI CEO Sam Altman is traveling to Washington this week to update the Trump administration on the company's latest and most advanced AI system. The briefing will highlight the AI's capabilities, including solving complex problems beyond human reach and identifying security vulnerabilities independently. The visit aims to accelerate governmental approval for this groundbreaking technology, Axios reports.
