Researchers using Twitter’s formerly free Decahose data service will soon be required to pay a fee of $42,000 per month to continue using the service and retain their data.
Decahose, a streaming service that gives scientists unfettered access to around 10% of all tweets in real-time, has been a mainstay for academic research on myriad topics, including emergency response, law enforcement-related activities, political misinformation, and extremism.
Earlier this year, due to Elon Musk’s purchase of the company, Twitter announced it would begin charging for access to its numerous APIs with fees ranging from $100 to $42,000 per month.
Academics and universities using the Decahose service would be required to pay the “enterprise” fee of $42,000 per month under the new rules and the amount of data available would reportedly drop from 10% of the total live tweets to 0.3%.
According to a report from British publication i news, researchers were recently sent an email explaining that they could either start paying for their access or delete any data they’ve obtained:
“Researchers who don’t sign the new contract ‘will need to expunge all Twitter data stored and cached in your systems.’ Researchers will be required to post screenshots ‘that showcase evidence of removal.’ They have been given 30 days after their agreement expires to complete the process.”
The timing of this change is noteworthy, as it comes mere days after Musk joined Florida Governor Ron DeSantis in announcing the latter’s bid for the White House in the 2024 U.S. presidential election.
If executed as planned, the impending change to Decahose access could stifle global research efforts to study election misinformation and social manipulation ahead of the 2024 election as it occurs on what’s been described as one of the world’s most politically-active social media platforms.
Beyond misinformation research, the pricing changes would also limit academia’s ability to study internet-related crimes such as human trafficking and financial scams. Furthermore, Decahose has served as one of the internet’s largest repositories of human sentiment — a data source that drives insights and predictions on nearly every society-related research topic imaginable.
Related: Elon Musk threatens Microsoft with lawsuit, claims AI trained on Twitter data
Much of the data used to study cryptocurrency sentiment, for example, comes from Twitter and Reddit data silos. By limiting access to this data, Twitter is potentially hindering both ongoing research and forestalling new efforts.
One explanation for the pricing increase could involve Musk’s ongoing efforts to ensure technology companies aren’t using Twitter’s data to train their AI systems.
Source: Read Full Article