How AI is transforming researchers’ daily work

Once a speculative technology, artificial intelligence is now embedded in researchers’ daily routines, reshaping how they spend their time and energy. Findings from Elsevier’s Researcher of the Future report show where AI is making the biggest difference and reveal what researchers expect and need from the technology to drive further adoption. AI is helping researchers reclaim time Rising demands and limited resources are putting increasing pressure on researchers.

• Only 45% of survey respondents feel they have enough time for research

• 68% say the pressure to publish has increased in the past two to three years.

• Funding uncertainty adds to the strain, with just 33% expecting more funding in their field soon.

Despite these challenges, researchers remain committed to producing high-quality, trustworthy work. In that context, they value AI for protecting time for activities such as study design, analysis, collaboration and mentoring.

• 58% percent say AI already saves them time

• 69% expect it to use it for those tasks in the next two to three years.

Researchers use AI a to sustain research quality under pressure, rather than as a replacement for established practices, and they are putting it to use in specific ways.

Researchers are using AI to manage information overload

The 2025 Researcher of the Future report indicates that 58% of researchers have already used AI tools for work, up from 37% the previous year. Primarily their focus is on using AI to manage information overload.

• 61% use AI to find and summarize new studies.

• 51% use it for literature reviews.

These tasks are time-consuming and complex, especially at the start of projects. Researchers view AI as an assistant, not an author. While adoption is growing, human expertise remains central to framing questions, evaluating quality and generating insight. Researchers see much of AI’s value in streamlining routine, information-heavy tasks, freeing allowing them to focus on deeper analysis and creative thinking.

Some use AI to assist with more complex, time-consuming tasks

A smaller percentage of researchers use AI for more sophisticated tasks, including some that see them competing with other research teams for funding.

• 41% use AI to draft grant proposals, streamlining a process that is often time-consuming and competitive.

• 38% rely on AI to analyze research data, accelerating insights and enabling researchers to handle larger, more complex datasets.

• 38% use AI to write or draft research papers and reports, helping to improve clarity, structure, and efficiency.

As adoption grows, researchers expect AI to play an even greater role in shaping how research is conducted, communicated, and funded, but certain requirements need to be met.

Further adoption depends on trust

The optimism around the deepening assistance AI can provide is tempered by concerns about training and oversight. Forty-five percent feel undertrained in using AI, and only 32% rate their institution’s AI governance as good. Trust, not just technical capability, shapes adoption. Researchers are clear about what would help build confidence:

• 59% percent trust AI tools more when references are automatically cited

• 55% value systems trained the most up-to-date scholarly literature

• Expectations for accuracy, transparency and oversight reflect longstanding research norms and a focus on integrity.

According to the report researchers emphasize the need for trust markers to strengthen confidence in research and AI tools alike:

• Transparency and clear citations (59%)

• Recency of data and inclusion of up-to-date literature (55%)

• Training on high-quality, peer-reviewed content (54%)

• Regular human validation of AI outputs (49%)