What have fact‑checking organizations documented about trends in political misinformation since 2016?

Checked on December 19, 2025
Disclaimer: Factually can make mistakes. Please verify important information or breaking news. Learn more.

This fact-check may be outdated. Consider refreshing it to get the most current information.

Executive summary

Fact‑checking organizations and researchers report a clear expansion and professionalization of the field since 2016, accompanied by a decisive reorientation from policing elite political claims toward rapidly debunking viral social‑media misinformation — a shift driven by the 2016 “fake news” shock and platform partnerships such as Meta’s third‑party fact‑checking program (3PFC) [1] [2]. Empirical studies find mixed outcomes: some measures show a relative decline in engagement with identified “fake news” on Facebook after platform interventions while other platforms and novel threats — notably AI‑generated content — create fresh challenges for fact‑checkers [3] [4] [5].

1. Rapid growth and professionalization of fact‑checking

Independent and newsroom‑linked fact‑checking has grown quickly: scholars and the Duke Reporters’ Lab document a surge in organizations worldwide since 2016, with hundreds of nonpartisan fact‑checkers operating in dozens of countries by the early 2020s — a movement that transformed from a niche journalistic practice into a global field with shared standards such as ClaimReview [2] [1] [6].

2. A field pivoted from elite claims to policing viral hoaxes

Multiple studies and practitioner accounts record a pronounced “debunking turn”: leading fact‑checkers moved resources away from methodical checks of political elites toward rapid response to trending hoaxes, memes, and conspiracies on social platforms — a reorientation reinforced by financial and operational partnerships with platforms that pay or route viral items to fact‑check partners [1] [7].

3. Platforms acted; measured diffusion trends changed but unevenly

Researchers using engagement data conclude that platform interventions correlate with a sharp fall in measured interactions with identified fake‑news sites on Facebook after 2016, while interactions on Twitter rose or remained elevated, producing a shifting platform landscape rather than an overall eradication of misinformation [3] [4] [8]. At the same time, authors stress limits to these measures: lists of “fake” sites and platform flagging miss many false claims and private channels, so trend estimates are partial [4] [9].

4. Local and regional fact‑checking expanded but remains outmatched

State and local fact‑checking initiatives multiplied in U.S. election cycles, with dozens of locally focused projects producing hundreds of checks, yet practitioners and the Reporters’ Lab warn these efforts are vastly outnumbered by the volume of claims circulated by campaigns and social networks — a capacity mismatch that local initiatives cannot fully close [10].

5. New tools, automation and the arrival of AI changed practice

Fact‑checkers adopted computational tools and archives (Hoaxy, ClaimReview metadata, NLP pipelines) to surface and verify claims at scale, and the rapid rise of generative AI since the early 2020s introduced a new wave of synthetic political content that fact‑checkers are only beginning to measure and counter with AI‑assisted workflows [11] [5].

6. Consistency, partisanship perceptions, and critique of the field

Meta‑analyses of fact‑checker verdicts show both substantial agreement across major outlets on many political claims and notable differences in coverage and emphasis, while public opinion research finds partisan gaps in whether fact‑checking is seen as fair — Republicans more likely to view fact‑checking as biased — a dynamic that fuels political attacks on fact‑checking partnerships and sometimes shapes funding and editorial choices [7] [6] [12].

7. What fact‑checkers document — and what remains uncertain

Fact‑checkers and scholars have documented the field’s growth, the shift toward debunking, platform policy effects, and new technical threats, but they also repeatedly caveat their claims: measurements rely on imperfect lists, platform data is incomplete, private messaging and multimedia misinformation are under‑observed, and causal links between exposure, belief, and behavior remain contested — areas that fact‑checkers flag as gaps in the evidence base [4] [9] [1].

Want to dive deeper?
How have Facebook's and Twitter/X's policy changes since 2016 differentially affected the spread of flagged misinformation?
What methods and tools (like Hoaxy or ClaimReview) do fact‑checkers use to scale verification, and what are their limits?
How is generative AI changing the kinds of political falsehoods fact‑checkers encounter and the verification techniques they deploy?