AI Newsletter January 2026

Welcome to the Dixon AI Newsletter which summarises selected AI stories and releases you might have missed. This issue covers AI news from 16th December 2025 to 23rd January 2026. The AI Macro Environment and Enterprise Insights

• Apple-Google AI Partnership Reshapes Mobile AI Landscape: Apple confirmed on 12 January that Google's Gemini will power the next generation of Apple Foundation Models, including a revamped Siri. The multi-year deal, reportedly worth around $1 billion annually, signals Apple's strategic pivot away from building its own frontier models and consolidates Google's position as an AI infrastructure provider to multiple ecosystems. This partnership may reshape enterprise expectations around default AI assistants on mobile devices. See (blog.google). Link: https://blog.google/company-news/inside-google/company-announcements/joint-statement-google-apple/
  

• OpenAI to Introduce Advertising in ChatGPT: OpenAI announced on 16 January that it will begin testing ads in ChatGPT for free and lower-tier users in the US. Ads will appear at the bottom of responses and will be clearly labelled. The company pledged that ads will not influence ChatGPT's answers and that user data will not be sold to advertisers. Premium subscribers remain ad-free. The move brings OpenAI's business model closer to Google and Meta and signals how AI companies may monetise large free user bases. See (openai.com). Link: https://openai.com/index/our-approach-to-advertising-and-expanding-access/
  

• US-China Chip Tensions Escalate with H200 Restrictions: Despite US government clearance for export, China blocked Nvidia's H200 AI chips in January, prompting suppliers to halt production. The move intensifies the semiconductor decoupling between the two economies and may accelerate enterprise efforts to diversify supply chains and evaluate alternative compute architectures. See (theguardian.com). Link: https://www.theguardian.com/technology/2026/jan/17/china-blocks-nvidia-h200-ai-chips-that-us-government-cleared-for-export-report
  

• AI Jobs Growing Amid Broader Hiring Weakness: The share of US job postings mentioning AI reached a record 4.2% in December 2025, even as overall hiring remained flat. Nearly 45% of data and analytics postings now contain AI-related terms. However, entry-level tech hiring continues to contract, with reports suggesting a 25% decline at major tech firms since 2023. Workforce planning should account for both rising AI skill demands and the changing shape of entry-level roles. See (hiringlab.org). Link: https://www.hiringlab.org/2026/01/22/january-labor-market-update-jobs-mentioning-ai-are-growing-amid-broader-hiring-weakness/
  

• Microsoft Spending with Anthropic Set to Reach $500 Million: Reporting indicates Microsoft is on track to spend $500 million with Anthropic, intensifying competition with its existing OpenAI partnership. The move reflects hyperscaler strategies to maintain optionality across multiple frontier model providers and may influence enterprise procurement approaches to AI platform selection. See (theinformation.com). Link: https://www.pymnts.com/artificial-intelligence-2/2026/microsoft-spending-on-anthropic-approaches-500-million-a-year/ 
  

• Davos 2026: Energy, Social Licence and the Shift from Hype to Scale: AI dominated discussions at the World Economic Forum, with leaders pivoting from capability hype to implementation challenges. Microsoft CEO Satya Nadella warned that AI risks losing "social permission" to consume energy unless it delivers tangible improvements in health, education and productivity. He framed energy costs as the key determinant of which economies win the AI race, describing compute tokens as a new global commodity. White House AI adviser David Sacks reinforced the administration's push against state-level regulation, warning of 1,200 AI bills in US state legislatures. See (cnbc.com). Link: https://www.cnbc.com/2026/01/20/microsoft-nadella-ai-race-energy-tokens.html
  

• Hassabis and Amodei on AGI, Jobs and China: In a rare joint appearance at Davos, Google DeepMind CEO Demis Hassabis and Anthropic CEO Dario Amodei offered converging warnings despite being competitors. Amodei maintained his forecast of "Nobel Prize level" AI by 2026-27, while Hassabis held to a 50% probability of AGI by 2030, citing "missing ingredients" around scientific creativity. Both predicted significant near-term disruption to entry-level white-collar roles, with Amodei reiterating that up to half could disappear within five years. Both expressed a preference for slower development to give society time to adapt. See (weforum.org). Link: https://www.weforum.org/stories/2026/01/live-from-davos-2026-what-to-know-on-day-2/
  

New and updated AI model and application releases

• OpenAI 

• GPT Image 1.5: Released 16th December 2025 with improved instruction-following, more precise editing capabilities and up to 4x faster generation speeds. https://openai.com/index/the-new-chatgpt-images-is-here
  

• GPT-5.2-Codex: Released 18th December 2025, optimised for agentic coding with improvements on long-horizon work through context compaction, stronger performance on large code changes and enhanced cybersecurity capabilities. https://openai.com/index/introducing-gpt-5-2-codex
  

• Google
  

  • Gemini 3 Flash: Released 17th December 2025, offering frontier intelligence built for speed with Gemini 3 Pro-grade reasoning at Flash-level latency and cost. Now the default model in the Gemini app and AI Mode in Search. https://blog.google/products/gemini/gemini-3-flash/
    

• Microsoft
  

  • No updates from this company this month.
    

• Anthropic
  

  • Agent Skills ecosystem expansion: Announced 18th December 2025 with organisation-wide management for Team and Enterprise plans, a directory of partner-built skills, and Agent Skills published as an open standard for cross-platform portability. https://anthropic.com/news/skills
    

  • Cowork: Announced 12th January 2026 as Claude Code for non-technical users, extending agentic capabilities beyond coding to general work tasks. https://anthropic.com/news/claude-code-on-the-web
    

  • Claude's Constitution: Anthropic released a comprehensive 23,000-word constitution for Claude, significantly expanding from the 2,700-word 2023 version. The document explains the reasoning behind Claude's values and behaviour rather than simply listing rules, covering four core priorities: broadly safe, broadly ethical, compliant with Anthropic's guidelines, and genuinely helpful. The constitution addresses complex topics including Claude's potential consciousness and moral status, and is already being used in Claude's model training process. Released under a Creative Commons CC0 1.0 Deed, it can be freely used by other AI developers. Claude's new constitution \ Anthropic. Also see the full constitution text.
    

• Notable Other Model Releases
  

  • Grok Voice Agent API: Released 22nd December 2025 - enables developers to build voice agents that speak dozens of languages, with an average time-to-first-audio of less than 1 second. https://x.ai/news/grok-voice-agent-api
    

  • Mistral OCR 3: Released 17th December 2025 - achieving new frontier for accuracy and efficiency in document processing. https://mistral.ai/news/mistral-ocr-3
    

• Notable Other AI Application News
  

  • Manus Acquisition: Meta acquired Manus, the Singapore-based AI agent company behind general-purpose AI automation technology, in a deal reportedly worth $2 billion. Manus Joins Meta to Expand General AI Agent Deployment. The acquisition is currently facing regulatory scrutiny in China over potential technology export control violations. Beijing intensifies scrutiny of Meta's $2 billion Manus acquisition
    

  • Manus 1.6: Manus released version 1.6 featuring significant architectural upgrades for autonomous task execution. Key new capabilities include mobile application development for iOS and Android, and Design View, an interactive visual canvas for point-and-click image editing. Introducing Manus 1.6: Max Performance, Mobile Dev, and Design View
    

Recommended Reading, Listening and Book Releases

• For anybody:
  

  • Video: Google's Demis Hassabis, Anthropic's Dario Amodei Debate the World After AGI at Davos: 

  • FULL DISCUSSION: Google's Demis Hassabis, Anthropic's Dario Amodei Debate the World After AGI | AI1G

  • Podcast: AI Daily Briefs: An accessible daily podcast for keeping up with all the latest AI news.

  • Book: The AI Transformation Playbook by Rob Dixon was published in December 2025 in both paperback and ebook formats. The AI Transformation Playbook
    

From Dixon AI

• AI Drop In: one-hour, live session held three times a week. Join here: https://www.dixonai.com/ai-drop-in 

  • w/c 26th Jan: What’s New in AI

  • w/c 2nd Feb: Copilot - what’s new and tips

  • w/c 9th Feb: Exploring Claude Code
    

• New Website: We launched a new website in December at www.dixonai.com. Take a look and let us know what you think!
  

A word from Rob

As we start 2026, one thing has become very clear. There is a widening split in who is actually getting a return on investment from AI.

On one side, the big tech firms. Microsoft, Google, NVIDIA, Tesla. They are embedding AI deeply into software development, engineering, and product design. The result is dramatic productivity gains, major efficiency improvements, and entirely new products hitting the market at speed.

On the other side sits what the tech world rather unkindly calls “legacy industry”. In practice, this means almost everyone else. Marketing, manufacturing, engineering, construction, healthcare, insurance, law, transport, finance. The backbone of the real economy.

Most of these organisations tried something with AI in 2025. Pilots, tools, experiments. But the uncomfortable truth is that the vast majority are not seeing meaningful ROI. The figures coming out of firms like McKinsey suggest that somewhere between 80 and 90 percent of AI initiatives outside big tech are failing to deliver material impact on the bottom line.

The reason is not the technology.

It is capability.

Too many organisations have treated AI as a tool problem rather than a people, process, and capability problem. They have not invested fast enough in upskilling, shared understanding, or new ways of working. Without that foundation, AI remains interesting but economically ineffective.

What changes in 2026 is pressure.

This is the year when boards, investors, and markets stop accepting “we’re experimenting” as an answer. AI-driven ROI will move from optional to existential.

There is also a darker edge to this. I recently listened to a fascinating All-In Podcast episode from early January, featuring Jason Calacanis in conversation with Bob Sternfels from McKinsey and Hemant Taneja from General Catalyst. They discussed a very real private equity play: acquiring legacy businesses in sectors like insurance, healthcare, construction, and manufacturing, stripping them back, and rebuilding them as AI-first organisations.

That is disruption, not theory.

So the real question for 2026 is this. Will existing organisations develop AI capability fast enough to become the disruptors in their own industries, or will new entrants do it to them?

At Dixon AI, we believe speed is everything. That is why I wrote The AI Transformation Playbook. It sets out, step by step, how organisations can build AI capability at pace and turn experimentation into impact.

This year will answer a big question.

Will transformation come from within, or will it arrive from the outside?

Either way, AI’s impact on the bottom line is no longer a future concern. In 2026, it becomes a test of survival.

Rob

Dixon, R.J. (2025) The AI Transformation Playbook: A step-by-step guide to becoming an AI-enabled organisation. London: Dixon AI Press. Available at: https://www.amazon.co.uk/dp/1919400508 (Accessed: 26 January 2026).

All-In Podcast (2026) Why AI will dwarf every tech revolution before it: robots, manufacturing, AR glasses from CES 2026. Presented by Chamath Palihapitiya, Jason Calacanis, David Sacks and David Friedberg, 8 January. Apple Podcasts. Available at: https://podcasts.apple.com/us/podcast/why-ai-will-dwarf-every-tech-revolution-before-it-robots/id1502871393?i=1000744237064 (Accessed: 26 January 2026).