My understanding is that the fraud claim — if it matches the fraud claim in the New York case, at least – is this. Exxon had an internal figure used to represent one kind of climate change costs for one purpose. Investors never saw this. There was also an external figure shared with investors, to help investors assess risks to Exxon stock. The case for fraud is that these weren't different for a good reason, but because they sought to mislead investors.

The pricier of these figures is the external figure, so they're being charged for being too alarmist with the public. In either case I don't think it has much to do with the previously linked research claims in this thread.

(To this, there is added some consumer fraud by... not labelling gas pumps with warnings about global warnings, among other things, I think it was? This is the weaker claim, and I would not expect it to prevail.)