The Solutions Journalism Network (SJN) has received a grant from The Rockefeller Foundation that will enable SJN and Meedan to build a first-of-its-kind tool that leverages new generative large language models (LLMs) to rapidly evaluate and classify news stories using SJN’s criteria for rigorous solutions journalism. Although the tool will first be used to identify climate solutions stories, it will be architected so that it can ultimately be deployed for any issue area.
Importantly, this tool will enable SJN and others to more efficiently identify patterns, insights and themes that are emerging from responses to social and environmental challenges that can help inform social innovation worldwide, something SJN is already working to do through its Insights Lab platforms.
SJN and Meedan look forward to engaging with other journalists and journalism support organizations, technologists, researchers and philanthropists to ensure that our efforts contribute to that of others seeking ways to use AI tools to better inform the public about how communities are responding to climate change and forging promising paths forward.
“The AI classifier that Meedan is developing will make it possible for people to search specifically for news stories that show potential to respond to real-time challenges related to climate change,” says SJN CEO David Bornstein. “By making it possible to systematically identify rigorous reporting about climate adaptation, mitigation and resilience efforts from far and wide, this tool promises to speed up the global learning needed to make headway against the climate crisis.”
“Our development process will produce key design recommendations and data sets which can be reused by the community independently, or in collaboration, with the appropriate open licensing,” says Meedan CEO Ed Bice. “To the extent feasible, Meedan will operate open-source practices and transparent, repeatable evaluation processes which help ground the conversation and define which benchmarks we have established.”
SJN, an independent nonprofit organization, is leading a global shift in journalism, focused on what the news misses most often: how people are trying to solve problems and what we can learn from their successes or failures.
- Online conversations are heavily influenced by news coverage, like the 2022 Supreme Court decision on abortion. The relationship is less clear between big breaking news and specific increases in online misinformation.
- The tweets analyzed were a random sample qualitatively coded as “misinformation” or “not misinformation” by two qualitative coders trained in public health and internet studies.
- This method used Twitter’s historical search API
- The peak was a significant outlier compared to days before it using Grubbs' test for outliers for Chemical Abortion (p<0.2 for the decision; p<0.003 for the leak) and Herbal Abortion (p<0.001 for the decision and leak).
- All our searches were case insensitive and could match substrings; so, “revers” matches “reverse”, “reversal”, etc.