So we have a mountain of insecure code -- backdoors and no-ops created by Opus (https://news.ycombinator.com/item?id=48520661 https://news.ycombinator.com/item?

So we have a mountain of insecure code -- backdoors and no-ops created by Opus (https://news.ycombinator.com/item?id=48520661 https://news.ycombinator.com/item?id=48520661) -- are they saying they're not going to let Fable fix it? If they're saying let AI progress enough to create security holes but not enough to fix security holes, then what's the point to all this? Has the AI coding model reached its self-imposed limit?