1. Microsoft shifts in AI Enterprise Sales: Microsoft is doubling down on integrating AI Agents into its AI enterprise strategy, with the creation of the CoreAI division. The unit is led by former Facebook executive Jay Parikh. He reports directly to CEO Satya Nadella. The effort focuses on streamlining AI enterprise operations and delivering AI-driven productivity tools using Agents. Besides the organizational changes, Microsoft is also experimenting with variable usage based pricing vs traditional seat based pricing. These mirror similar efforts by Salesforce for its Agentforce AI services. These efforts underscore Microsoft’s ambition to lead in enterprise AI transformation, now leveraging external senior management talent, a departure from its traditional management growth from inside strategy.. More here.

   

2. AI Hallucinations persist: The AI industry continues to wrestle with the persistent issue of AI-generated "hallucinations"—instances where AI systems produce false or misleading information. Despite robust, ongoing efforts by LLM AI companies like OpenAi, Anthropic et al to mitigate this problem, hallucinations remain a fundamental challenge in generative AI. This is partly due to the inherent design of large language models (LLMs), which relies on generating stochastic matrix math predictions via ‘intelligence tokens’, to predict results based on patterns in training data. And this problem continues through the various Scaling innovations on the AI inference side. This is leading to plausible but inaccurate outputs. Recent research seems to suggest that these errors are an unavoidable aspect of current and new AI architectures. This results in an ongoing tension between the rapid advancement of AI technologies and the need for accuracy and reliability. Currently the drive for innovation often overshadows concerns about misinformation. And customers seem to be taking it in stride for now. All this underscores the importance of human oversight and the development of strategies to manage and reduce the impact of AI hallucinations. Also this raises the possibility that this may be a ‘Forever Problem’ with LLM AIs. More here.

   

3. Google’s opportunities with ‘proactive spinoffs’: As Google faces at least two major antitrust actions, one in the remedies phase, analysts are exploring the opportunity around spinoffs for the company. The reports game out Google spinning off assets like Google Cloud, YouTube et al. The resulting ‘sum of the parts’ analysis suggest valuations meaningfully higher than current levels. And point to historical spinoffs like AT&T that led to net positive outcomes for investors. Google could potentially and proactively accelerate innovation by creating independent entities focused on specific AI-driven opportunities. This approach could unlock value, foster agility, and allow for more targeted development in areas like cloud computing, advertising, and other emerging technologies. More here.

   

4. Rising concerns over AI ‘Breaking Internet’: There is rising concern that AI could meaningful impact the internet, as it alters the SEO Search driven ad economics of the current internet. Current moves by LLM AI companies especially to drive AI powered Search means more AI summaries and AI Agents by Google with Gemini, OpenAI, Perplexity and others, which in turn potentially reduce the outgoing traffic to internet content and web publishers who rely on that traffic for further monetization. Thus the concerns over an AI driven ‘broken internet’. The offset of course will be new monetization mechanisms once the new AI Search innovations are more prevalent, particularly with the other wave by leading LLM AI companies to innovate around ‘AI Browsers’. In particular, Google, OpenAI, Perplexity, and others are actively focused on innovating around AI Browsers. The two issues are concurrent in nature, and potentially offer some solutions to a rising short-term dislocation. Also, a video podcast discussion of this subject from the perspective of Gen Z and older generations here. More here.

   

5. Anthropic’s Amodei highlights white-collar AI job losses: Recent warnings from Anthropic CEO Dario Amodei regarding the potential impact of AI on the job market, particularly on the ‘white collar’ side. Amodei predicts that AI could eliminate up to 50% of entry-level white-collar jobs within the next five years, potentially increasing unemployment rates to 10–20%. He emphasizes that sectors such as technology, finance, law, and consulting are particularly vulnerable to AI-driven disruption. Amodei criticizes both the government and the tech industry for downplaying these risks and urges for greater transparency and proactive measures to address the impending changes over time. The comments revive ongoing concerns over the broader and longer term implications of AI advancements, noting that while AI offers significant benefits and net positive job impact, it also poses serious challenges to employment and societal structures.. More here.

   

Other AI Readings for weekend:

1. Nvidia Jensen Huang’s tough balancing act between US and China. More here.

   

2. Huawei’s short term challenges are ramping up against Nvidia and others. More here.

   

Up next, the Sunday ‘The Bigger Picture’ tomorrow. Stay tuned.

(NOTE: The discussions here are for information purposes only, and not meant as investment advice at any time. Thanks for joining us here)