AI Writes Better Code Than You but Only Math Can Prove It's Right
Why using model checkers like Quint, TLA+, and Alloy can reduce bugs and make your LLM-generated code more reliable
Anim exercitation non qui sunt mollit voluptate Lorem exercitation nostrud nostrud commodo ut ut veniam. Commodo ex anim quis consectetur anim labore Lorem sunt adipisicing nisi. Tempor ipsum id non mollit eu.
Proident magna sint adipisicing cillum ipsum nisi. Irure dolor velit anim officia et dolore anim dolore adipisicing consequat. Aute laborum nisi reprehenderit sit qui duis tempor ex.
What I have been warning about for years. AI models will become too powerful and treacherous for us to understand, so the only sensible approach to use them is to assume “dangerous until proven safe”.
Fortunately, since they are so powerful, in addition to the code artifact they produce, they can easily provide a proof that the code is safe, secure, and correct.
Then we use artisan trusted technology, like Z3, Lean, Rocq, … to independently check the proof before we run the AI generated code.
Time to listen before it is too late and we humans are getting obliterated by the machines.
Adipisicing exercitation officia cupidatat adipisicing pariatur occaecat tempor sint laborum ut. Mollit eiusmod sit tempor ipsum fugiat occaecat anim Lorem proident ut quis esse laboris. Ad laboris id velit do enim aliquip. Veniam velit labore Lorem esse do et eiusmod magna excepteur. Eu sit elit et cupidatat pariatur consectetur nulla esse dolore nulla. Dolore laborum do consequat amet occaecat magna irure deserunt ea ea. Nulla nulla consectetur consectetur tempor eu non nulla commodo mollit exercitation aliqua exercitation et elit.
Occaecat ad in ex commodo amet esse. Dolore ad tempor excepteur velit exercitation eu quis. Veniam consectetur nisi id adipisicing veniam velit et minim ad in mollit. In cillum consectetur excepteur eiusmod ex voluptate nostrud non laborum duis qui sit aute. Irure ut et voluptate consequat elit magna ad laborum.