how do we know the ChatGPT models haven’t crawled the publicly accessible breach forums where private data is known to leak? I imagine the crawler models would have some ‘follow webpage-attachments and then crawl’ function. surely they have crawled all sorts of leaked data online but also genuine question bc i haven’t done any previous research.

At one point while reviewing his own handwritten notes, McConney acknowledged his memory was “incorrect” about a specific disclosure on Trump’s statement. Seeing the handwritten edits also published in the final statement version, he acknowledged he must have primarily written the disclosure about asset valuations that deviated from generally accepted accounting principles.

someone get this fraud a golden raspberry