According to Deloitte’s, enterprise use of AI agents is on the rise, with 25% of enterprises using general AI forecast to deploy AI agents by 2025, increasing to 50% by 2027. 2025 Global Forecast Report. Focusing on the impact of Gen AI on the technology, media and telecommunications industries, Deloitte’s report defines autonomy Innovative AI agents — also known as agent AI — is a software solution that can complete complex tasks and meet goals with little or no human supervision.
Also: Why Agentic AI is the new power and nearly 80% of business leaders are afraid of the dark
Here are some of Deloitte’s more interesting predictions about synthetic and agent AI:
- The acceptance gap among women in using Gen AI is rapidly closing: By 2025, women’s GenAI testing and usage is expected to equal or exceed men’s, but tech companies still need to improve trust, representation in training models, and diversity. form in the AI workforce. By 2023, women’s Gen AI usage rate will be half that of men.
- Gen AI is driving data center energy consumption to skyrocket: Global data center power consumption is forecast to double to 4% (1,065 terawatt hours) by 2030 as AI generation power consumption grows faster than other uses. and other applications.
- Gen AI is set to make devices smarter: By 2025, the percentage of Gen AI-enabled smartphones shipped could exceed 30%, with about 50% of laptops capable of handling Gen AI locally.
A closer look at the AI agent adoption forecast from the report notes the following: “The growth of AI agents will be driven by innovation from both startups and leaders Established industry identifies new revenue opportunities Built on large language models, these AI agents will offer greater flexibility and more use cases than learning methods. machine or traditional deep learning While the ultimate goal is to achieve the autonomous and trusted agents, Deloitte expects significant improvements in their capabilities by 2025 as these technologies develop rapidly, with AI moving past pilots and evidence of concept in some markets and for some applications by 2025.”
Also: Agentic AI is the leading strategic technology trend in 2025
According to Deloitte, agent AI has commonalities characteristics and capabilities:
- Built on the model foundation: Platform models like LLM enable AI agents to reason, analyze, and adapt to complex and unpredictable workflows, making them more flexible than RPA and expert systems. LLM is improving rapidly, with enhanced reasoning abilities and the ability to break tasks into smaller steps among the most recent breakthroughs.
- Autonomous activities: Although the degree of autonomy varies, AI agents can be trained to independently plan and execute complex tasks. By introducing reasoning tokens, thought chain models can solve more complex challenges than previous LLMs.
- Feeling the environment: AI agents can perceive the environment, process information, and understand the context of assigned tasks. Advanced agent AI can process multimodal data, such as video, images, audio, text, and numbers.
- Use tools: Agentic AI interacts with tools and systems to complete tasks, such as software, enterprise applications, and the internet.
- Orchestra: Agentic AI can direct the engagement of other systems and bots to complete a task. For multi-agent systems, this means collaborating with other autonomous generative AI agents.
- Memory access: LLM is stateless. Each interaction is processed independently, and information is not retained once the interaction is complete. With the addition of database and retrieval mechanisms, AI agents can access short-term memory to maintain context while performing a specific task and long-term memory to learn and improve. improve experience.
Also: Businesses are grappling with what to do with Gen AI, venture capitalists say
According to Deloitte, Gen AI chatbots and co-workers are highly sophisticated; they can interact intuitively with humans, synthesize complex information, and create content. However, they lack the level of autonomy and autonomy that agent AI promises.
Deloitte highlights the key differences between copilots, chatbots and AI agents. According to Deloitte’s definition, “Agent AI has ‘agency’ — the ability to act and choose what action to take. Agency implies autonomy, that is, the power to act and make decisions independence. When we extend these concepts to agentive AI, we can say that it can act on its own to plan, execute and achieve goals — it becomes an ‘agent’. Goals are set by humans, but actors determine how those goals are realized.”
Sales forceAI Connected Customer Status of study revealed that one-third of consumers prefer working with AI agents for faster service. Many consumers are happy to communicate with AI agents but also want to know when that conversation takes place.
Also: Sticker shock: Are businesses increasingly disillusioned with AI?
Besides, Agent AI is the top strategic technology trend for 2025 according to Gartner. Companies will invest heavily in AI agents as the world of work changes forever. Research shows that an executive is the key to unlocking value from these conversions. The rapid adoption of general AI and agent AI is creating great pressure on IT organizations. CIOs must also serve as chief AI officers, according to one Sales force survey. While the majority of CIOs believe AI is a game-changer, only 11% say they have fully deployed the technology, citing data infrastructure and security as their top hurdles .
To learn more about Deloitte’s general AI and agent AI future trends, you can visit This.
