Welcome to 2026, the year AI continues its methodical journey toward pragmatism.
There will be no AI bubble in 2026. ROI problem for AIregardless of how people interpreted MIT’s “State of AI in Business 2025” report on return on investment. (How do you measure the ROI of proofs of concept, and why would you measure the ROI of POCs?)
As a reminder, AI is still in its infancy, and this is particularly true for enterprise AI. The majority of companies have less than three years of experience with AI, meaning that AI structures, culture, processes, and lifecycle management are not yet fully developed, let alone mature, in most organizations.
Those who succeeded with application of AI — for example, companies like Adobe, Salesforce, ServiceNow, and SAP — make up a relatively small group of AI innovators who have one thing in common: They started with AI five to ten years ago, committed to the technology as a long-term investment. These organizations learned during this time what was needed in terms of structure, culture, processes and lifecycle management. When generative AI and agentic AI materialized, they adapted quickly and accelerated.
All of this is to say that research from Omdia, a division of Informa TechTarget, repeatedly shows that companies are engaged in AI. In 2026, organizations will accelerate their investments and commitments, continuing the seemingly boring but essential work of implementing AI in their organizations.
Is this a prediction? No, it’s a trend. This has been around for a few years and will continue to be around for a few more years. Let’s move on to some predictions.
Free ChatGPT shuts down or is severely restricted; users move to Gemini, Copilot
OpenAI cannot keep ChatGPT free unless it moves to an ad-supported revenue model. It’s not likely, otherwise it would have happened by now. Google and Microsoft see their Gemini and Copilot apps through a different lens; they are largely complementary to adjacent revenue streams, such as search and Microsoft 365. As such, they can offer these apps for free or at reduced rates. Once OpenAI starts losing ChatGPT users, what happens? It will lose its market influence and cease to dominate AI stock trends.
Physical AI emerges
In 2026, a market ecosystem comes together to accelerate the adoption of a new domain: physical AI. HCL Technologies has a definition that works for our purposes: “At its core, physics AI unites cognition and mechanics. It combines AI models, simulation and real-time perception with robotics and sensors to create systems that learn from experience, adapt to uncertainty and work seamlessly with humans. …Traditional AI operates in the digital domain, generating predictions or automating software workflows. roles towards autonomous operational capability.
At CES this week, NVIDIA announced the availability of a full physics AI stackincluding robot foundation models, simulation tools and state-of-the-art hardware.
Physical AI is not just about robots for manufacturing and logistics; it also has applications in industries such as energy, transportation, construction, public safety, and general field service management.
Sovereign AI will trigger AI innovation in Europe and India
Sovereign AI refers to the ability of a specific nation or region to maintain autonomous control over its AI ecosystem, including data governance, AI model development, infrastructure ownership, and regulatory frameworks. These include ensuring domestic AI capabilities, protecting jurisdictional interests, and reducing dependence on foreign AI technologies and services.
The concept emphasizes national autonomy in AI capabilities, often for reasons of data sovereignty, security, resilience and economic competitiveness.
Corporate adoption of sovereign AI in the EU, UK and India will see significant growth in 2026. As these markets are large enough and have self-sustaining addressable markets, there is a good chance that domestic AI ecosystems will gain traction and displace some US and China-based vendors. This shift is unlikely to happen with GPUs or CPUs, but rather with services and software – deep AI models in languages other than English come to mind. To some extent, the smaller the market size (Denmark, Finland, Slovenia, etc.), the better the chances for a regional supplier to thrive. When businesses have access to models that are better suited to their language, the results will theoretically be more accurate, thereby accelerating the time to production of AI initiatives.
Furthermore, India represents a massive and self-sustaining market opportunity, rivaling the US and Chinese markets. Sovereign AI regulations in this country can boost potential players in the Indian AI ecosystem, giving them the opportunity to innovate and thrive in India and perhaps expand to challenge AI activities outside India.
Mark Beccue is a principal analyst at Omdia, where he covers artificial intelligence.
Omdia is a division of Informa TechTarget. Its analysts maintain business relationships with technology providers.