Was their only evidence the screenshot they included? It’s seems more likely that Google could have been looking for phrases like “Let’s look at it” rather than the term “Zoom”.
I would give you that if Google was a small company without much experience in the field. But they're, what, the largest tech company that exists? There is no way this was approved without MANY conversations about the end user experience and what this actually does/is perceived.
At some point, benefit of the doubt goes out the window. And I would argue that a company the size of google should no longer get the benefit of the doubt.
But, I really dislike how Google often gets a free pass (from the tech community) on these things, and frankly, I prefer the sensationalized reality over our more accurate one because it rightfully puts pressure on them to stop messing up so often.
How often does Google get to say "whoopsie daisy, our silly AI messed up again" before we need to start asking them serious questions like "why is there an AI making recommendations in the first place when those recommendations have such a high probability of preferring Google products"?
You could even create intentional situations where you know how the algorithm will behave because you know the data you are training on but still have the option to chalk up mistakes to 'accidental'.