Not true at all. In fact settlements mostly happen because it would cost significantly more for a company to go through discovery and argue their case in court regardless of result. And court systems strongly encourage this to save their own time. There an entire industry of patent trolls and sleazy personal injury lawyers in business because of this.

Not sure about that, don't defendants sometimes settle because they don't want the publicity of a trial or don't want their dirty laundry being aired in discovery?

Not always. Sometimes it’s as simple as: settling is cheaper than proving you’re “clear” at trial.

I was sued. I was 19 years old working as a painter for a dishonest contractor that paid crap wages. I nosed out of a parking lot after work one day to see around a line of cars turning in and a big sedan ploughed into my little econobox. Several years later, as the statute of limitations was about to run out, the driver of the sedan sued me. My insurance companies first move, before doing any discovery, was to offer her $50k. She said no, so discovery began. It turned out she'd been mis-prescribed an anti-psychotic to create the symptoms she was suing me for having caused. The case was thrown out. The insurance company's legal bills ended up being much less than $50k, but the way it worked was they took a guess at the break even point, offered a bit less than that, and made an offer.

That's not to say this is how it works when Meta is on trial. I just thought it was useful perspective on the nature of settlements.

She's not wrong. The discovery process has shown that such decisions were made by Meta and Zuck himself, knowingly, in the face of research that opposed their goals.