Andrew Bailey, in his capacity as the chair of the Financial Stability Board, has expressed concern over the potential destabilising effects of advanced AI technologies on the global financial system. He highlighted that cutting-edge AI models are demonstrating increased autonomy, sophisticated problem-solving skills, and potential threat capabilities, raising alarms about their impact on economic stability. This warning was conveyed in a detailed letter to international finance ministers and central bank governors at the G20 meeting.
Nvidia has informed its largest clients that the prices of servers equipped with its AI chips will increase by over 15%, effective early next year. This price rise is largely attributed to increased memory costs. The company is set to release its quarterly earnings report next week. Customers are now facing the direct financial impact of the ongoing memory shortage on AI server procurement.
Nvidia's recent research reveals that the effectiveness of AI agents relies heavily on fine-tuning and integration methods rather than just the underlying AI model. Even AI models with less-than-ideal performance can achieve strong results when properly managed and adjusted.
Businesses are shifting their preferences as each AI lab launches new models, creating volatility that raises questions about the loyalty and long-term commitment of enterprise AI spending for investors in both firms.
For years, the tech world lived by "if the product is free, you are the product." We accepted this as the norm. However, after numerous data scandals and privacy missteps, this mindset is evolving. We're now witnessing a significant change where privacy becomes a fundamental software feature rather than an afterthought.
Anthropic has introduced Claude Mythos 5 for code scanning through Claude Security, integrating it with partner defense products. Instead of direct model access, users receive the model’s findings as output. Alongside this, Anthropic is pledging $35 million in credits to support open-source security initiatives.
Agentic AI is shedding light on leadership deficiencies that have long hindered business growth, offering fewer excuses and more accountability for effective management.
After years of observing corporate AI adoption, it’s clear that selecting which AI model to use—Copilot, GPT, Gemini, Claude, Grok, or even Chinese models like Deepseek and Qwen—is often the first step companies ponder. However, this choice might not be as crucial as it seems. Think of the AI model like a microprocessor: important, but just one part of a larger system. Companies don't always need the most powerful model for every task; simpler models can handle routine queries cost-effectively, while more complex tasks are assigned to advanced models. This layered approach is already being adopted, with firms like Deepseek focusing on competitive pricing and Microsoft pushing smaller, specialized models for specific industries and tasks. The real edge lies beyond the model itself—in how companies leverage their unique institutional knowledge, workflows, documents, and feedback loops to build smarter, tailored AI systems. This creates a profound competitive advantage, as institutions with rich, context-specific learning outperform those relying solely on generic, large models. Ultimately, it’s about institutional sovereignty: owning your data and learning determines how valuable your AI implementation truly is, far beyond just picking the latest model.
Slack, owned by Salesforce, has launched Slack Code, a new feature that integrates AI coding agents like Anthropic's Claude Code, Cognition's Devin, GitHub Copilot, and Vercel's agent directly into dedicated Slack channels. This allows entire teams to collaboratively watch, guide, review, and ship software, moving AI coding out of individual terminals into a transparent, multiplayer workflow. Slack Code creates project-specific channels where all work, including code diffs and live previews, is visible and archived for audit. The approach shifts the bottleneck from coding itself to judgment and creativity, expanding participation across roles such as engineers, product managers, and designers. Slack emphasizes security by restricting agent permissions to those of the invoking user, ensuring no cross-team data leaks. This launch positions Slack as a key platform for collaborative AI-driven development amid Salesforce's strategic pivot toward AI-powered enterprise solutions. Leaders in the space see a future where multiplayer channels and single-user terminals coexist, with a growing preference for collaborative workflows fostering higher quality and efficiency in software development.
OpenAI, which had earlier opposed California's AI safety bill SB 53, is now advocating for the state to enhance the legislation to better address AI safety concerns.
