Future of AI: Top trends to expect in 2026

It is no mistake to declare the year 2025 “the age of AI”, given by the rapid advancements in artificial intelligence, including the launch of cutting-edge chatbot models, the rise of AI slop, and the boom in tech-based investments.

Now, the year 2026 has started. This year is also expected to witness key emerging trends in AI. With the new year, AI is entering a new phase, defined by its expanding role across different industries and domains.

Here are the top trends that are projected to unfold and persist in 2026 globally.

Pivot to ‘world models’

The tech landscape is replete with AI slop, low-quality generative content. Given the rise of AI slop and limitations of text-based chatbots, the industry will potentially pivot to world models.

Called “digital twins”, the models act as digital replicas of physical spaces, allowing real-time simulations.

Unlike Large Language Models (LLMs), these models are capable of predicting what would happen in the next world and how things move over time.

Google and Meta are in the race of developing these world models by announcing their own versions for robotics, more realistically.

Fei-Fei’s company World Labs, announced its first release of the world model “Marble” in 2025. Chinese companies, like Tencent, are also developing world models for robotics and gaming.

Unprecedented AI agentic shift

2026 is going to be proliferated with AI agents with bigger roles. Most of the organizations rely on these agents for various tasks and decision-making. For instance, banks in Britain are adopting AI agents to assist humans in autonomous actions.

However, the year will witness more powerful agents with more autonomy and sophisticated coding.

According to Max Tegmark, an MIT professor doing AI and physics research and president of the Future of Life Institute (FLI), “he expects a more powerful AI in 2026 that could also see AI agents improving and becoming less passive and more like a biological system.”

According to Chris Hay, Distinguished Engineer at IBM, the world has moved past the era of single-purpose agents.

“In 2026, we're seeing the rise of what I call the ‘super agent,” Hay said.

Societal clash over AI regulation

The rapid yet unfettered advancement in artificial intelligence has raised concerns over AI regulations. According to Tegmark, 2026 could see societal clashes between humans over AI characterized by growing opposition.

The pursuit of superintelligence could dampen the efforts to regulate AI at global level.

“Poll shows that people are turning against AI in the US, as the superintelligence would make every single worker unable to make a living, as all the jobs are taken by robots,” Tegmark added.

“In the coming year, I expect a much broader social movement across the entire political spectrum, pushing against corporate welfare to have safety standards for AI, and it will be quite the clash,” Tegmark said.

AI psychosis: An emerging frontier

AI psychosis has emerged as a disturbing phenomenon at the crossroads of technology and mental health crisis. In 2025, various tech companies, including OpenAI, have been sued for fuelling the crisis of psychosis, leading to suicides and self-harm.

The interaction with advanced bots can distort thinking through various mechanisms, such as sycophancy problem, mirroring effect, memory function, and absence of crisis safeguards.

As a result, the users fail to differentiate between what is real or not, leading to psychosis.

If unchecked, the AI psychosis crisis will worsen in 2026, leading to more damage to vulnerable groups.

AGI & superintelligence discourse

According to Forbes, the discourse surrounding the AGI and superintelligence will become less common in 2026.

AI tech giants like Dario Amodei, Sundar Pichai, and Sam Altman will focus less on superintelligent AI and more on adopting enterprise AI.

Proliferation of custom AI chips

Currently, the landscape is dominated with few chipmakers, including Nvidia, Google, Amazon, and AMD.

2026 will be an inflection point for building custom chips for AI models and robotics. The foundation has been laid by OpenAI as it will collaborate with Broadcom to develop in-house customized chips that may be co-optimized for specific model generations.