5 Generative AI Trends to Watch in 2025

Generative AI is as trendy as it has ever been.

This year, research in AI was awarded the Nobel Prize, and the world’s largest tech companies incorporated AI into as many products as possible. US government AI boosted As a driver and a strategic pillar for federal spending in building a clean-energy economy. But what’s next for 2025?

Generic AI trends in the last few months of 2024 point to greater pressure for adoption by tech companies. Meanwhile, results on whether AI products and processes see ROI for enterprise software buyers are mixed. While it’s difficult to predict how AI will continue to shape the tech industry, experts have made predictions based on current trends.

respondents of a ieee study In September AI was ranked as one of the top three areas of technology that will be most important in 2025 in 58% of cases. In contrast, almost all respondents (91%) agree that 2025 will see “a generic AI reckoning” in terms of what the technology can or should do. Expectations are high for generic AI, but the success of projects taking advantage of it remains uncertain.

1. AI agents will be the next hot topic

Based on my research and observations, the use of AI agents will increase in 2025.

AI agents are semi-autonomous generative AI that can chain together or interact with applications to carry out instructions in an unstructured environment. For example, Salesforce uses AI agents to call sales leads. Like generative AI, the definition of the agent’s capabilities is unclear. IBM defines it as an AI that can solve complex problems, such as OpenAI O1. However, not all products billed as AI agents can reason in this way.

Regardless of their capabilities, AI agents and their use cases will likely be at the forefront of generic AI marketing in 2025. AI “agents” may be the next step in the evolution of this year’s AI “copilots.” AI agents can spend time working through multi-stage jobs independently while their human counterparts handle other tasks.

2. AI will both help and hurt security teams

Cybersecurity attackers and defenders will both continue to leverage AI in 2025. 2024 has already seen the proliferation of generator AI security products. These products can write code, detect threats, answer prickly questions, or serve as “rubber ducks” for brainstorming.

But generative AI can present information that is inaccurate. Security professionals can spend the same amount of time double-checking the output as they would if they had done the work themselves. Failure to review such information can break code and lead to even greater security problems.

“As AI tools like ChatGPIT and Google Gemini become deeply integrated into business operations, the risk of accidental data exposure increases with new data privacy challenges,” Jeremy Fuchs, cybersecurity evangelist at Check Point Software Technologies, said in an email to TechRepublic. ” “In 2025, organizations must move quickly to impose tighter controls and governance over the use of AI, ensuring that the benefits of these technologies do not come at the expense of data privacy and security.”

Generative AI models are susceptible to malicious actors like any other software, especially through jailbreak attacks.

“The growing role of AI in cyber crime is undeniable,” Fuchs said. “By 2025, AI will not only increase the scale of attacks but also increase their sophistication. With AI continuously learning and adapting, phishing attacks will become harder to detect.

Generative AI could make traditional methods of identifying phishing emails – bad grammar or unexpected messages – obsolete. As the proliferation of AI-generated video, audio, and text increases, disinformation protection will become more important. As a result, security teams must adapt to both use and defense against generative AI – just as they have adapted to other significant shifts in business technology, such as mass migration to the cloud.

3. Businesses will evaluate whether AI provides ROI

“The pendulum has swung from ‘new AI innovation at any cost’ to a resounding imperative to prove ROI in board rooms around the world,” Uzi Dvir, global CIO of digital adoption platform company WalkMe, said in an email. “Similarly, employees are asking themselves whether it’s worth the time and effort to figure out how to use these new technologies for their specific roles.”

Organizations struggle to determine whether generative AI adds value and in which use cases it can make the most difference. Organizations adopting AI often face high costs and unclear goals. It can be difficult to measure the benefits of using generic AI, where those benefits manifest, and what to compare them to.

This challenge is a side effect of the integration of generic AI into many other applications. This leads some decision makers to wonder whether generative AI add-ons actually increase the value of those applications. AI levels can be expensive, and over the next year, expect more companies to rigorously test — and sometimes discard — features that don’t deliver results.

Many companies that are incorporating generative AI on a large scale are finding success. at its Q3 earnings callGoogle attributed this result to its AI infrastructure and products like AI Overview. However, Meta pointed out that there could be significant growth in AI capital expendituresEven if user numbers have declined.

WATCH: Google is previewing its sixth generation of cloud AI accelerator Trillium.

4. AI will have a big impact on scientific research

As well as impacting enterprise productivity, contemporary AI has seen significant movement in science.

Four of the 2024 Nobel Prize winners used AI:

• Demis Hassabis and John Jumper of Google DeepMind won Nobel Prize for Chemistry For protein structure prediction with AlphaFold2.
• John J. Hopefield and Geoffrey Hinton won Nobel Prize for Physics For his decades of work developing neural networks.

The White House held a summit on October 31 and November 1 Use of AI in life sciencesHighlighting how AI enables solutions to complex challenges in ways that impact the world. This trend is likely to continue next year as generic AI models develop and mature.

5. Environmental tools made from AI will not compensate for its energy losses

Energy efficiency is another buzzword in AI.

But for every use case in which AI can help predict weather patterns or optimize energy use, there’s another story about the environmental cost of building the data centers needed to run generative AI. Such construction requires enormous amounts of electricity and water – and rising global temperatures only exacerbate the problem. It is unlikely that any balance will be achieved in a problem of this scale.

However, for businesses, expect to see companies making dubious and anecdotal claims of energy savings and environmental friendliness around AI. Consider the resource usage associated with your organization’s AI strategy.

What are the most popular generic AI products?

The most famous generic AI products are:

• ChatGPT, OpenAI Chatbot
• google gemini
• Microsoft Copilot
• GPT-4, the big language model behind ChatGPT
• DALL-E 3, an image generator

What is the most advanced generative AI?

Various tests have been proposed as possible criteria for determining the most advanced generative AI. Some organizations rate their models on human educational standards, such as the International Mathematical Olympiad or Codeforces competitions.

Other assessments, such as measuring large-scale multitask language comprehension, were created explicitly for generative AI. Google’s Gemini Ultra, China Mobile’s Jiutian and OpenAI’s GPT-4o are on top mmlu leaderboard Today.