Elon Musk says he’ll make Twitter’s algorithm open source next week. As my colleague Christiaan Hetzner notes in his write-up of the development, Musk promises a lot of things and often fails to deliver, but, if the poly-CEO does follow through on this, it could be the most consequential decision of his Twitter tenure.
There would be three major effects of open-sourcing Twitter’s algorithm. The first would be political: Allowing outsiders to poke around the code should settle any questions about alleged ideological bias within the company’s algorithm itself.
That’s not quite the same issue as the historical liberal bias that was supposedly exposed in the “Twitter Files” episode, in which Musk granted a few journalists access to Twitter’s pre-Musk internal communications and let them craft a selective account of what went down in terms of human content-moderation decisions. But with the algorithm, everyone would get equal access, so it could prove more genuinely illuminating.
Secondly, if Musk really does mean “open source” in the traditional sense, Twitter’s algorithm would become available for modification and redistribution, by anyone. I still find it hard to believe he’s preparing to let people basically copy Twitter, but let’s see what the fine print says. (Open-source Twitter rival Mastodon must be watching carefully.)
The third effect would be the most far-reaching: The world would get to see what happens when Big Tech opens up one of its black boxes.
This could go horribly wrong. It’s possible that, when shown how Twitter’s systems make decisions, miscreants could swiftly learn how to game the system and inundate Twitter with (even more) spam and hate speech. That’s certainly the kind of outcome Big Tech has over the years cited as the main reason for keeping its black boxes closed.
However, as open-source advocates have been arguing for just as long, open code can make for better safety, because a host of outside experts get to pore over it and spot weaknesses that internal staffers missed.
If it turns out that the benefits of opening up an algorithm like Twitter’s outweigh the risks, there may be major implications for the future of artificial intelligence regulation.
In Europe, an Artificial Intelligence Act is being considered that would ensure that “high-risk A.I. systems shall be designed and developed in such a way to ensure that their operation is sufficiently transparent to enable users to interpret the system’s output and use it appropriately.” That would require a level of algorithmic transparency that Big Tech for now heavily resists.
If Musk makes good on his promise, that could be the first sign of the dam breaking. Everyone should be watching to see what happens next.
Want to send thoughts or suggestions to Data Sheet? Drop me a line here.
David Meyer
Data Sheet’s daily news section was written and curated by Andrea Guzman.
