Is ChatGPT losing its edge? Prof Alexiei Dingli on why the AI race is entering a new phase

For much of the past three years, it was difficult to imagine the future of artificial intelligence without ChatGPT at its centre. The tool that brought generative AI into the mainstream reshaped how individuals, businesses and institutions interact with information. Yet, according to Prof. Alexiei Dingli, that sense of inevitability is now fading.

In a recent blog post, Prof. Dingli argues that while ChatGPT remains one of the most widely used AI tools in the world, the environment that enabled its early dominance has fundamentally changed. “It once seemed impossible to imagine the future of Artificial Intelligence without ChatGPT at the centre… However, by the end of 2025, that dominance no longer feels certain,” he writes.

ChatGPT’s rise was unprecedented. Launched publicly in November 2022, it reached 100 million users within two months, making it the fastest-growing consumer application at the time. By mid-2025, OpenAI reported more than 700 million weekly active users. However, scale has not translated neatly into profitability.

“Running large language models requires expensive computing infrastructure,” Prof. Dingli notes, pointing to OpenAI’s heavy reliance on power-intensive GPUs supplied largely by Nvidia. Despite growing subscription and enterprise revenues, “high operating costs continue to limit profitability. As a result, more users do not always translate into more profit.”

This cost structure has become a strategic vulnerability. Nvidia still controls over 80 per cent of the AI accelerator market, meaning pricing or supply disruptions could directly impact OpenAI’s capacity. While rivals such as Google and Microsoft have invested heavily in custom chips, OpenAI’s own reported efforts in this area remain at an early stage.

Narrowing gaps and rising rivals

Beyond costs, the quality gap between AI models is shrinking. In 2023, GPT-4 was widely seen as the gold standard. By late 2025, however, competing models such as Google’s Gemini, Anthropic’s Claude and Meta’s Llama had reached comparable performance levels on many benchmarks.

“Public benchmark results and blind testing show that differences in quality are now often marginal,” Prof Dingli observes. As a result, users are increasingly prioritising price, reliability and ecosystem integration over raw model capability.

This shift has favoured platforms with deep hardware and software ecosystems. Google has embedded Gemini across Android and Workspace, Apple is rolling out on-device AI within iOS and macOS, and Microsoft is weaving Copilot throughout Windows and Office. By contrast, “OpenAI operates as a standalone platform… Without its own ecosystem, OpenAI is more vulnerable to changes in its partners’ platform strategies,” he writes.

The enterprise market presents another challenge. While OpenAI offers business subscriptions and APIs, large organisations often prefer providers with established compliance, security and deployment frameworks. Hyperscalers such as Microsoft, Google and AWS can offer cloud, on-premise and hybrid solutions alongside regulatory support – an area where OpenAI is still catching up.

At the same time, regulation is tightening. The EU’s AI Act and similar initiatives in the US are increasing compliance costs and raising barriers to entry. “Companies with large legal and regulatory teams are better positioned to adapt,” Prof. Dingli notes, suggesting that scale and institutional maturity will matter more than ever.

Perhaps most significantly, the industry itself is evolving. Rather than relying solely on massive, general-purpose models, businesses are increasingly adopting smaller, specialised and open-source systems that are cheaper to run and easier to deploy locally. “If the industry continues to move towards decentralised, modular AI, OpenAI’s model-centric approach could face added pressure,” he warns.

Developer behaviour is already reflecting this change. Once the default choice, OpenAI is now competing with a growing ecosystem of open-source and alternative models. Usage data from platforms such as Hugging Face point to rising adoption of local and custom deployments, reducing dependence on a single provider.

Market data reinforces this trend. Prof. Dingli highlights that in January 2026, Google’s Gemini gained significant market share, while traffic to ChatGPT declined year-on-year. “Although usage patterns vary, these data indicate that ChatGPT is rapidly losing its dominance in this domain,” he writes.

Crucially, Prof. Dingli is clear that this does not amount to failure. “None of this means ChatGPT is failing. It remains one of the most widely used AI tools worldwide,” he states. The challenge ahead is strategic rather than existential.

With OpenAI reportedly preparing for what could become a record-breaking IPO, expectations are rising. “OpenAI now faces a different kind of challenge, not survival, but reinvention,” Prof. Dingli concludes. Without meaningful progress in cost discipline, platform integration and enterprise readiness, ChatGPT risks becoming “a strong competitor rather than the defining face of the AI era.”

For businesses watching the AI landscape, the race is no longer about a single dominant tool, but about ecosystems, economics and long-term adaptability.