In a recently published blog post, Digital Science CEO Daniel Hook underlined Digital Science’s, Dimensions’ parent company, view on new technology, writing, “For any software development company, there is an implicit responsibility to the user communities that they serve…The rise of AI should raise the value that systems can bring to users, but it also raises the bar in the relationship between developer and user.” “We believe that research is the single most powerful tool that humanity possesses for the positive transformation of society and, as such, we have a responsibility to provide software that does not damage research.”

It is this view that underpins the launch of the new Dimensions solutions: Dimensions Research GPT and the Dimensions Research GPT Enterprise as well as the AI-driven summarization features

Grounding AI in scientific evidence

Dimensions Research GPT / Enterprise and the fully-integrated summarization feature have been developed with the view to better support the research community. These solutions combine the power of AI technologies with the robust scientific data available through Dimensions – the world’s largest collection of linked research data – to provide answers that are grounded in evidence.

Dimensions Research GPT is a tool connected to Dimensions data relating to tens of millions of Open Access publications, and is available for free for ChatGPT plus or Enterprise users, while Dimensions Research GPT Enterprise allows Dimensions users with a ChatGPT Enterprise account to draw from all Dimensions’ content.  Both versions allow researchers to access an advanced literature discovery workflow that merges the scientific evidence base of Dimensions with the Generative AI functionality of ChatGPT. These solutions mean that researchers can now access an advanced literature discovery workflow that merges the scientific evidence base of Dimensions with the Generative AI functionality of ChatGPT – reducing the likelihood of hallucinations and providing click-through scholarly references for each statement to enable quick and easy validation and further discovery. 

Our top tip for accurate responses: Effective prompts

“Effective prompts are crucial for obtaining accurate and relevant responses from both Dimensions Research GPT / Enterprise,” says Marta Ondresik, Senior Product Manager AI Solutions at Digital Science. “It is important to be as specific as possible about the topic, field or research along with any constraints or preferences,” she adds. 

Examples of ineffective prompts

  • Overly-broad questions such as “Tell me about science,”
  • Prompts lacking any specificity such as “What is the impact of science,”
  • Requesting access to restricted data.

Conversation starters for effective prompts

“To help users test Dimensions Research GPT we have created examples of effective prompts which are known as conversation starters,” explains Ondresik. Users can find these directly in the Dimensions Research GPT interface.

Examples of effective prompts

  •  Summarize the most common research methodologies used in recent studies on [topic], highlighting their strengths and weaknesses
  • What are the implications of quantum computing?
  • What are electromagnons?
  • Evaluate the current tools and technologies used in the field of [add field], including their limitations and potential for future development.

Accelerating interpretation

Meanwhile, Dimensions’ AI-driven summarization feature supports users in their discovery process for publications, grants, patents, and clinical trials, and addresses the daunting task of navigating vast volumes of information. The fully-integrated feature leverages AI to enable users to swiftly gauge the relevance of the research records featured in their search results. Whether utilizing the free web app or the enhanced Dimensions platform, the tool has the potential to accelerate discovery workflows and support interpretation of scientific data. 

Empowering the research community

While AI expands the potential of science, it also raises a host of concerns, from ‘paper mills’ churning out fake research to the misuse of data analysis known as p-hacking.  And that is why both the above-mentioned Dimensions solutions have been developed with scientific integrity and trust at the core.  In line with Dimensions’ commitment to introduce AI responsibly, both the new offerings aim to:

  • Help overcome challenges of sheer volume of content available, time-consuming tasks required in research workflows and need for trustworthy AI products
  • Provide users with the ability to contextualize the output they get so that they do not have to trust without verification.
  • Ensure that no groups of researchers are artificially, or through commercial approaches, disenfranchised or excluded from accessing this type of technology.

Dimensions’ latest solutions underscore a belief in harnessing the power of AI to empower the entire research community, while maintaining Digital Science’s wider commitment to responsible innovation.

Check out the following if you are interested in reading more about the topic:

If you would like to have more information, please contact the Dimensions team.