Google Says No Ads Planned for Gemini

Google DeepMind CEO Demis Hassabis has stated that the technology giant currently has no plans to introduce advertising into its Gemini chatbot, distinguishing its strategy from competitors like OpenAI who are exploring ad-supported models. This strategic clarification comes amidst a volatile week for AI infrastructure and enterprise adoption, with reports indicating Meta is pivoting away from custom silicon and OpenAI securing a major integration deal with ServiceNow.

Key Points

• Gemini Strategy: Google executives deny current plans for ads in the Gemini app, framing it as a distinct "assistant" tool separate from Search.
• Meta's Pivot: Reports suggest Meta is deprioritizing its in-house chip program, favoring immediate purchases of AMD processors to meet compute demands.
• Enterprise Expansion: OpenAI has signed a three-year deal to embed agentic AI directly into ServiceNow's IT workflows.
• Hardware Roadmap: OpenAI signals its dedicated AI device is "on track," with executives pointing toward a potential unveiling in 2026.

Google Clarifies Stance on AI Monetization

In a recent conversation at Davos, DeepMind CEO Demis Hassabis addressed growing speculation regarding the monetization of Google’s flagship AI model. Despite OpenAI’s recent moves to explore advertising within ChatGPT, Hassabis confirmed that Google is not currently developing an ad-supported tier for Gemini. He noted that while competitors might feel pressure to generate immediate revenue, Google’s capitalization allows for a different strategic pacing.

This statement appears to contradict earlier reporting from Ad Week in December, which cited anonymous clients claiming Google had targeted a 2026 rollout for Gemini ads. However, Google VP of Global Ads Dan Taylor reinforced Hassabis’s position, drawing a sharp line between the company’s core products.

"Search and Gemini are complementary tools with different roles. While they both use AI, Search is where you go for information on the web... and Gemini is your AI assistant. Search is helping you discover new information... We see Gemini as helping you create, analyze, and complete that."

While the standalone Gemini app remains ad-free for now, convergence is already visible elsewhere. Google continues to integrate commercial features into its AI Search experience, such as "direct offers" that present personalized discounts.

Meta Shifts Infrastructure Strategy

Significant changes are reportedly underway regarding how major tech companies power their AI ambitions. According to a research note from analyst Jeff Pu of Haitong Securities, Meta is deprioritizing the deployment of its custom silicon. Instead of relying on proprietary chips or becoming a primary customer for Google’s TPUs, Meta is reportedly placing substantial orders for AMD’s latest processors.

This pivot suggests a broader trend where "hyperscalers" are prioritizing immediate compute availability over the long-term goal of self-sufficiency. While Meta, OpenAI, and Anthropic all launched custom silicon programs to reduce reliance on Nvidia, the accelerating demand for compute power may be rendering these internal timelines obsolete.

Recent analysis suggests that established players like AMD now offer total cost of ownership and performance metrics that internal teams struggle to match. Consequently, Meta appears to be utilizing AMD chips to meet short-term requirements while potentially reserving custom silicon for highly specialized future workloads.

OpenAI Expands Enterprise Footprint and Hardware Ambitions

In the enterprise sector, OpenAI has finalized a three-year partnership with ServiceNow to integrate its models directly into the platform. This deal represents a shift toward "agentic" workflows, where AI does not merely generate text but actively executes tasks.

ServiceNow President Amit Zavery highlighted that these agents would be granted access to IT functions, such as remotely restarting computers or retrieving data from legacy mainframe systems. This integration aims to transform AI from a passive tool into an automated IT support teammate.

"Enterprises want OpenAI intelligence applied directly into ServiceNow workflows. Looking ahead, customers are especially interested in agentic and multimodal experiences so they can work with AI like a true teammate." — Brad Lightcap, OpenAI COO

Simultaneously, OpenAI is refining the timeline for its first hardware product. Speaking at the World Economic Forum, OpenAI Chief Global Affairs Officer Chris Lehane indicated the company is "on track" to unveil a device, potentially in the latter part of 2026. While details on the form factor remain guarded, Lehane pushed back against assumptions of an imminent 2025 release, emphasizing that the project remains in a development phase.

As 2024 progresses, the industry expects to see a surge in "land grab" tactics for enterprise contracts, alongside critical experiments to determine if agentic AI models can successfully transition from novelty to core business infrastructure.