J. Nathan Matias 🦣
banner
natematias.social.coop.ap.brid.gy
J. Nathan Matias 🦣
@natematias.social.coop.ap.brid.gy
Social & computer scientist who works alongside communities on science for a safer, fairer, more understanding Internet.

Founder, Citizens and Technology Lab · […]

[bridged from https://social.coop/@natematias on the fediverse by https://fed.brid.gy/ ]
@grimalkina tell me it’s going to be okay and that we’re not all just teetering on the ice of overoptimism about what we can measure?
November 29, 2025 at 6:34 PM
@ben <3
November 28, 2025 at 2:08 AM
In 2023, scientists estimate that 5.4k acute deaths and over 64k chronic deaths in North America were caused by wildfire smoke. Millions now face a question I ask every day: when is it safe to go outside?

AI forecasts might help, but is it safe to use them? […]

[Original post on social.coop]
November 25, 2025 at 1:24 AM
How dumb does a person need to think people are in order to send journalists public evidence that they are lying?

I guess that's what we can expect from someone on a team that scammed the postal service for tens of millions of dollars.

https://www.espn.com/olympics/cycling/news/story?id=6019436
Documents reveal USPS spending on Armstrong
As a grand jury investigates Lance Armstrong The File digs up previously undisclosed details about the U.S. Postal Service's sponsorship of his cycling team.
www.espn.com
November 20, 2025 at 11:27 PM
This guy told Velo News that Lance Armstrong, who was banned from professional cycling, had nothing to do with this new team.

Then Velo News noticed that Lance Armstrong was a sponsor of the team, and that his logo was on the official jersey.

Not a great start for someone claiming that honesty […]
Original post on social.coop
social.coop
November 20, 2025 at 11:24 PM
Reposted by J. Nathan Matias 🦣
Our new article in PNAS summarizes:

- Why is AI so hard to evaluate?
- Models of participatory science
- Five problems with participatory science
- Five areas where participatory methods can improve the science of AI

We hope this article helps researchers, communities, funders, organizers, and […]
Original post on social.coop
social.coop
November 17, 2025 at 12:38 PM
If you're an organizer, technologist, or scientist (of any kind) interested in participatory science about AI, I am hoping to continue a conversation we've already been having across @transparenttech and am considering a virtual meet-up in January.

Sign up here to express interest […]
Original post on social.coop
social.coop
November 17, 2025 at 1:08 PM
In September when @pluralistic spoke with my class, we asked him what he would suggest students do for a better technology industry after joining the workforce.

His answer? Join a union, because collective organizing is the bedrock of change.

Our article in PNAS picks up from there by asking […]
Original post on social.coop
social.coop
November 17, 2025 at 12:41 PM
Our new article in PNAS summarizes:

- Why is AI so hard to evaluate?
- Models of participatory science
- Five problems with participatory science
- Five areas where participatory methods can improve the science of AI

We hope this article helps researchers, communities, funders, organizers, and […]
Original post on social.coop
social.coop
November 17, 2025 at 12:38 PM