Kitchen's Ink

In the “not surprising except when it is” file, Google just updated its privacy policy to say it can dig into your social media posts – yes, yours – in order to inform its Artificial Intelligence products, including Bard.

According to Gizmodo, Google now “reserves the right to scrape just about everything you post online.”

Cool, cool.

This might be news to you, and you wouldn’t be alone. Bard, Google’s AI tool, doesn’t think this is happening yet, either. I found that out when I asked it on July 11.

“I do not do any social media scraping for my information gathering. I am trained on a massive dataset of text and code, which includes text from social media platforms. However, I do not actively scrape social media platforms for new information,” Bard began, before explaining more.

“There are a few reasons why I do not do social media scraping. First, it is against the terms of service of most social media platforms. Second, it can be difficult to get accurate and reliable information from social media. Third, social media scraping can be time-consuming and computationally expensive.

“Instead of social media scraping, I rely on a variety of other sources for information, including news articles, government documents, academic papers, and books. These sources are more likely to be accurate and reliable, and they are easier to access.

“I believe that it is important to be transparent about my information gathering methods. I want you to know that I am not using social media scraping to gather information. I believe that this is important for maintaining your trust and for ensuring that you are getting accurate and reliable information.”

I see. So… a few things stand out:

1. Despite the change in privacy policy, Google still, as of July 11, wants users to believe it isn’t actually doing the thing its new policy permits.

2. Bard repeated some iteration of “I am not using social media scraping” three times in a relatively long’ish explainer. It reminded me of a kid who knows they’re getting caught but wants to throw all the words at the adult as a deflector.

3. I feel even more strongly about the deflection thought as Bard told me it does not “actively scrape” and then went into … reasons … that included how time consuming and “computationally expensive” it would be.

If there are any companies I don’t buy time and resource excuses from – especially if that company feels it might gain a competitive advantage in a super-tight information arms race – it’s Google.

In any case, the surprising part is hearing/reading this blatant strategy out loud, right? I mean, we all *should* know (and if you didn’t, now you’re about to know) that AI tools really are only about as good as the input they’re given (even if they process it in different ways and do it faster). We figured that out almost immediately with ChatGPT 3’s inability to retrieve information past 2021.

In that light, give yourself enough time to think about things and you realize that the info compilation race had to happen (and has been with nearly every update since GPT 3) in order for those same AI tools to catch up as quickly as possible. That’s the not surprising part.

Whether the social media audience and profile builds from a decade ago, political forecasting built on connected information points, or now AI, it always, always comes back to one thing – data.

And yes, as we continue to pour ourselves into our social media lives, those feeds continue to serve as incredibly rich data sources for those attempting to make sense of, organize, and monetize the masses at any given time.

Don’t fear it, but just know everything you’ve said in messages and to your groups likely is going into “the machine.”