Prolific announces £25m funding to improve AI systems with high quality, verifiable human insights
Press release from Wednesday 12 July, 2023, London.
Prolific, a bootstrapped UK scale up founded to transform research by providing quick and easy access to verified and fairly paid participants, today announced it has raised £25m in a round co-led by global tech investment specialist Partech and Oxford Science Enterprises (OSE). The company has seen strong demand for its technology platform from organisations looking to use human insights and feedback to improve and train their AI models. This builds on the existing use of Prolific by academic and commercial researchers around the world to connect with a large and unrivalled pool of vetted respondents.
The new funding will help Prolific broaden its AI offerings and establish a stronger presence in the United States, home to over half of its current customer base, including industry giants like Google and prestigious institutions such as Stanford University. Prolific’s UK and European clients include the University of Oxford, King's College London, and the European Commission.
Prolific was founded in 2014 out of a frustration with how hard it was to find reliable and genuine participants for online research. Existing tools were slow, expensive and participants were not always who they said they were. Prolific set out to change this and built a sophisticated research infrastructure applying technology to the sourcing, pre screening, vetting, and paying of participants, as well as integrating research tools. Today, Prolific offers convenient access to over 120,000 active participants, enabling insights to be gathered swiftly from diverse and specialised groups.
For participants, Prolific provides a way to take part in fun, interesting and world-changing research, as well as earning extra money. Prolific is committed to fair compensation and has paid out over $100m to participants to date. AI will undoubtedly transform many areas of society and Prolific offers the opportunity for anyone to play a role in the safe development of this exciting new technology.
As AI models and applications grow exponentially, the source of the data used to build and train these systems becomes incredibly important and feeds into issues emerging around accuracy, bias, and copyright. Prolific’s platform matches AI developers with engaged and verified people (not bots) to help improve and fine tune their AI applications, helping to address these concerns:
- Inaccuracy - ‘Hallucinations’, or confident but incorrect responses by AI systems, have the potential to derail public trust in the technology and dissuade businesses from adoption. Prolific enables a process called Reinforcement Learning by Human Feedback (RLHF) where the output from AI models is checked by human respondents to improve and train them. For example, the outputs from a translation model could be checked by native speakers to make it more natural and authentic. The global data collection and labelling market (incl RLHF) is estimated to be a $17bn market by 2030.
- Bias - When it comes to data annotation, research has found that the demographics and backgrounds of annotators matters. With Prolific’s diverse and verified participants, AI models can be trained on truly representative population samples, reducing bias.
- Copyright – scraped or stolen data used to build or train AI applications is becoming an increasing concern. Prolific’s platform ensures clear auditing of sourced data, providing transparency that aligns with regulations proposed by the European Parliament.
The Series A funding will help Prolific further develop its AI proposition, including by further building highly specialised research groups which will be accessible via its platform. For instance, software developers (accessible via Prolific) can provide feedback on code generated by AI, improving its capabilities.
Phelim Bradley, CEO and co-Founder, Prolific, comments: “I’m incredibly proud of what Prolific has achieved already. Our platform is providing trusted insights to enable world changing research, and we’re excited to bring these advantages to new markets. AI represents one of the biggest leaps forward in technology in recent years and our unique approach to data sourcing from humans positions us to make these systems more accountable and less biased. The funding we have secured will fuel our growth in the AI space, especially in the US, bolstering our commitment to human-guided AI development during this pivotal moment in the technology's progression.”
“Prolific has built an incredibly powerful online platform for research,” said Omri Benayoun, General Partner at Partech. “Its roots in academia means that it has applied the highest standards to quality, while its technical expertise brings innovation that sets it apart from anything else out there. Where others rely on manual recruitment methods, Prolific has built a research infrastructure covering everything from the recruiting and vetting of participants to integration of research tools. Prolific is poised to conquer global leadership in academia and is also perfectly placed to aid the development of AI.”
Sam Harman, Investor Director, Oxford Science Enterprises, said, “We invest in companies with the potential to revolutionise their field and become market leaders, and that is why we’re incredibly proud to back the team at Prolific on their latest fundraise. Prolific is changing the way we build data-driven ecosystems, by empowering both researchers and participants alike, and we are excited to continue supporting the team on realising their mission of making trustworthy data accessible to everybody.”
Over 22,000 researchers have been active on the platform in the last year and a study is launched on Prolific every 3 minutes. The company, which now has more than 100 employees, became profitable in 2020 and has since reinvested all capital into its growth.
Contact: press@prolific.com
