Prolific

In July, our product team shipped several updates designed to take your data collection further. 🔧 - Active in the last X days filter: Target only active participants without manually pruning your groups. - Demographic breakdowns of participant groups: Analyze any group by age, marital status, and more directly in the Overview tab. - Configurable task layout in AI Task Builder: Define exactly how your stimulus, instructions, and content blocks are arranged. - Expanded verified expert network: Access a fast-growing pool of verified healthcare professionals and more, including 1,200 new language specialists added every month. - AND/OR logic in saved filter sets: Apply complex targeting conditions to a filter set and reuse them across studies. Amazon MTurk also closes to new customers on July 30. First-time Prolific researchers can redeem a $50/£50 coupon until then. Full roundup and coupon details ➡️ https://lnkd.in/eGcT_haU

Our VP of Data & AI Enzo Blindow says data volume is a solved problem. Quality is what separates a good AI model from a great one, and that changes everything about how labs source their training data. He sat down with Liam Lawson on The AI Report podcast to talk through what actually goes into high quality training data, why models lean too hard into stereotypes, plus the research we published at ICLR on commercial pressure in AI models. Full episode here → https://lnkd.in/emPMnj65

If you're evaluating research platforms right now, here’s what it looks like when you set up on Prolific. No sales call or contract needed. Create an account, connect your preferred tool (like Qualtrics, SurveyMonkey, Gorilla, your own API), access 300K+ verified participants and select from 300+ screeners, set your pay, and publish. Right now, you can get $50/£50 credit toward your first study. Claim it here 👉 https://lnkd.in/eDb-N9gb

During [ICML] Int'l Conference on Machine Learning week in Seoul, we brought together senior AI leaders for an exclusive dinner at N Seoul Tower 🇰🇷 Guests traveled up Namsan Mountain via the famous cable car before taking in iconic views of the Seoul skyline, then sat down for an evening of exceptional cuisine with researchers and practitioners from OpenAI, Microsoft AI, Google DeepMind, Ubisoft, and more - joined by our own AI research team, who brought fresh insights from their spotlight papers presented at the conference. Conversations spanned evaluation, alignment, and the role of human data in building models that actually work. Thanks to everyone who joined us. We'll be back with more!

The research industry has had a participant experience problem for years. I wanted to build something different with Prolific, where participant experience would be a priority. So I’m delighted to see that in Conjointly’s comparison of 11 platforms, we came out top on participant satisfaction, earnings, and enagement: - 98% of Prolific participants rated the platform easy to navigate - 96% said they'd recommend it - Prolific participants also reported the highest median monthly earnings Participant experience is inseparable from data quality -- how can you expect to get the data you need if the experience the end users have is not excellent. Full study in comments.

When you're measuring how a whole country feels about AI, sample quality matters as much as the questions you ask. This OII study by Dr Florence Enock and Professor Helen Margetts on UK trust and use of LLMs surveyed a nationally representative sample of 2,002 UK adults, sourced via Prolific. We're pleased to see the findings now publicly available. You can access their report below 📄

📊 Our research team members Oriol Bosch Jover and Andrew Gordon ran a snap poll on the Farage Clacton by-election story, surveying a representative sample of 1,174 UK adults weighted to match the UK population on age, sex, region, education, housing tenure, and 2024 vote. Farage says he's standing again so voters can judge him directly. The public isn't convinced, only 25% believe that's his real reason, and 73% think he probably did something wrong regarding the gifts and support under investigation. Reform UK's own voters see it differently. 84% believe his explanation, only 17% think he did anything wrong, and most say a win would clear his name outright. Every other major party has chosen not to contest the by-election, but 69% of the public, including a majority of Reform voters, think they should. Check out the full breakdown and methodology: https://lnkd.in/eCPtb53C

New snap poll from our research team on the Farage Clacton by-election story. A representative sample of 1,174 UK adults, weighted to match the UK population on age, sex, region, education, housing tenure, and 2024 vote. Full results and blog from Oriol below 👇