Skip to main content
Anthony Falco

Anthony Falco

Chef and author

Jul 30, 2026

3:36
Mm-hmm.
3:36
These automations, especially automations that are self-iterating, where you can run something for, you know, these more powerful models that are coming out, they're, they're gonna run for, or they can run for hours on a process and, or a project.
3:49
And all of those activities, all of the, the, all of the, the entire AI pipeline or LLM pipeline is generating logs.
3:57
Where did this data come from? How has it changed? Who did the training? Um, what, actually, what resources were used to, to run, uh, to, to run that, uh, that, to build that model? Uh, you know, just all the metadata about the actual creation of some sort of an artifact from, from LLMs in addition to the data provenance, how that data was mutated.
4:20
It's, it's an enormous amount of work that is being done and being recorded in logs and, uh, and, and, and there's a lot of signal in there that's being generated that includes the humans that are either the subject of or the actors in creating that.
4:35
The prompts, the, the qu- you know, the prompts especially that, that show intent, the way that data was used, you know, what data sources were brought in that might involve personal information.
4:45
It's just an enormous amount of privacy, uh, data with high privacy implications that is buried in this avalanche of data.

11 MINS LATER

15:46
So it's, you essentially focus on what we are talking about, but I'm more curious on how do you make sure that that's within the premise of, I would say, privacy principles or laws?

We value your privacy

We use cookies to understand how you use our platform and to improve your experience. Click “Accept All” to consent, or “Decline non-essential” to opt out of non-essential cookies. Read our Privacy Policy.