Stay in the know: AI trends and early predictions for 2026

What is agentic AI, and why does it matter?

Agentic AI refers to autonomous, intelligent systems that can adapt to changing environments, make complex decisions, and collaborate with other agents and humans. It’s AI that enables organizations to automate not just repetitive tasks, but also dynamic, multistep processes. Agentic AI is poised to improve efficiency, unlock new business models, and free up human talent for higher-value work.

How can agentic AI be used?

Among the many potential applications for agentic AI are:

• Customer service: AI agents can triage and resolve support tickets, escalating only the most complex cases to humans.
• Supply chains: Agents can autonomously optimize inventory, logistics, and procurement in real time.
• Finance: Automated portfolio management, fraud detection, and regulatory compliance monitoring can be handled by AI agents.

What are people saying about agentic AI?

Agentic AI adoption is still in the early stage. The majority of AI leaders and representatives surveyed indicated that their organizations are either in the pilot phase with limited deployment of AI agents or have not deployed agents at all. Very few have had a full-scale rollout of agentic AI currently. Those that noted moderate or significant deployment of agents generally represented larger organizations in tech-focused industries.

However, public sentiment is optimistic about the near-term implementation of agentic AI. Nearly half of LinkedIn respondents believed autonomous AI agents will significantly transform their organizations in the next two to three years. Less than 5% of LinkedIn poll respondents believed that agentic AI will have no impact on their organization in this timeframe, with the remaining participants responding that agents will drive limited to moderate impact.

Three early predictions for 2026

1. Scaling pilots to production: Agentic AI will likely move beyond pilot projects and become more widely adopted across industries, especially among larger organizations equipped with the necessary capital and talent. The exploding market of “out-of-the box” agentic solutions for common use cases will expand usage of agents to more industries.

2.  Increased focus on governance and compliance: Organizations will have to prioritize and implement well-defined governance frameworks and usage guidelines to address the unique considerations and risks of autonomous agents.

3.  Agentic-focused upskilling and reskilling: Organizations will also begin exploring and investing in training programs that help employees adapt to the new ways of working and new roles, including “agent ops” teams responsible for monitoring, training, and governing AI agents.

What is physical AI, and why does it matter?

Physical AI embeds intelligence into the physical world, enabling machines to interact with their environment in meaningful ways. It integrates artificial intelligence with robotics, autonomous vehicles, the Internet of Things (IoT), and digital twins to sense, interpret, and act in the physical world. Examples include everything from warehouse robots to smart medical devices and traffic lights. Physical AI can unlock new efficiencies and improve safety in sectors where automation was previously limited by complexity or cost.

What are some uses for physical AI?

Physical AI applications are transforming industries:

• Manufacturing: Robots and AI-powered quality control systems reduce defects and downtime.
• Logistics: Autonomous vehicles and drones streamline deliveries and warehouse operations.
• Health care: Wearable devices and smart sensors enable real-time patient monitoring and adaptive therapies.

What are people saying about physical AI?

The majority of AI leaders surveyed predicted minimal to moderate usage of physical AI in the next two to three years within their organizations. The speed of adoption of physical AI technology is highly dependent on sizable barriers, including stringent safety and security requirements, as well as the substantial costs associated with hardware deployment and ongoing maintenance. Additionally, challenges such as regulatory compliance, integration with existing infrastructure, workforce readiness, and public acceptance further impact the widespread implementation of physical AI solutions.

General population LinkedIn respondents indicated more varied opinions about how physical AI may impact company operations in the next two to three years. More than a third of respondents expected minimal impact from Physical AI. However, more than 50% expected moderate to significant impact. Just over 10% of respondents did not expect physical AI to affect their organization’s operations at all. Skepticism over the impact of this technology is likely shrinking. Early impacts of physical AI are apt to affect some industries faster than others as well.

Three early predictions for 2026

1.  Paced adoption: Physical AI could gain significant traction first in asset-heavy, task-intensive sectors—manufacturing, logistics, health care, and agriculture—while remaining less common in many knowledge- or service-led industries. Adoption is likely to grow quickly where the ROI is compelling and environments are structured but may proceed cautiously in industries where work is principally digital, highly interpersonal, or constrained by security, privacy, and brand-experience considerations.

2.  Focused on safety and security: Organizations will be smart to prioritize safety and security measures to mitigate the risks associated with physical AI. Those adopting this technology should pair simple physical safeguards (emergency-stop buttons, light curtains that halt machinery if people get too close, collision sensors) with fail-safe software, strong cyber defenses, and clear audit trails so physical AI runs without putting people, assets, or data at risk.

3. Harmonization of human-machine collaboration: Employees and physical AI agents will begin learning how to more seamlessly operate as integrated crews, enabling workers to focus on higher-value tasks and improve overall productivity. To accelerate this shift, companies should invest in intuitive interfaces, reskilling programs, and targeted change-management practices.

What is sovereign AI, and why does it matter?

With data privacy regulations tightening worldwide, organizations must ensure compliance while maintaining operational flexibility. Sovereign AI helps ensure that data, model weights, and compute resources remain within specific national or regional boundaries. Along with addressing regulatory, privacy, and geopolitical concerns, sovereign AI can help build confidence with customers and partners and reduce dependency on foreign technology providers.

What is the potential impact of sovereign AI?

The implications and impact of sovereign AI are far reaching:

• Health care: Patient data can be processed and stored locally to comply with privacy laws.
• Finance: Transaction data and AI models can remain within national borders to comply with regulations.
• Public sector: Government AI systems can be designed to improve transparency and local control.

What are people saying about sovereign AI?

AI leaders agreed that data residency constraints and compute considerations are important to strategic planning for their organization—though the level of importance of sovereign AI varies across industries. The urgency rises significantly in highly regulated sectors, such as banking and insurance, life sciences and health care, energy and industrials, and telecommunications. In these sectors, strict rules, highly sensitive information, critical-infrastructure status, IP considerations, and even national security concerns make local control over data and models imperative.

Unsurprisingly the broader LinkedIn community also viewed sovereign AI as a top strategic consideration. With a steady stream of headlines—ranging from cross-border data-privacy disputes and high-profile cyberattacks to looming EU regulations—there are clear signals that data controls and compute infrastructure are rapidly becoming board-level concerns.

Three early predictions for 2026

1. Increased regulatory scrutiny: Governments will continue to introduce new regulations governing data privacy, security, and AI governance. With the varied levels of regulatory scrutiny growing globally, the outlook is more uncertain in the US as early-stage bills develop. Organizations should proactively implement robust compliance programs—encompassing transparency, explainability, and continuous monitoring—not only to avoid legal penalties and protect brand trust, but also to preempt regulation.

2. Initial demand for sovereign AI solutions: Organizations will begin to seek out AI solutions that comply with local laws and regulations. To localize data and compute resources, organizations are expected to adopt multi-cloud and edge computing strategies. Those that can successfully navigate sovereignty requirements will undoubtedly gain customer confidence and access to new markets.

3. Rise of regional and national AI hubs: Countries and regions may invest in building local AI ecosystems to foster innovation and economic growth within their borders. These hubs aim to attract top talent and capital while reducing reliance on foreign AI capabilities.

These surveys reflect excitement and enthusiasm about agentic AI and physical AI and recognize the importance of sovereign AI to strategic planning. Increased efficiency, productivity, and innovation are among the likely results of these trends along with more robust guardrails to safeguard data and its usage worldwide. We’ll continue to monitor these and other emerging trends to share updated predictions for the coming year in future posts.

Want to learn more?

To find out the latest research, market trends, and viewpoints on innovative technologies that are shaping our future, subscribe to the Deloitte AI Institute newsletter and follow the AI Institute for future blog posts in this series.