Why Sorry Dave?
Safe AI begins with Safer Code
In 2001: A Space Odyssey, the main character is outside the mothership, asking the AI to "open the pod bay doors HAL." In a pleasant response, the computer replies, "I am sorry, Dave, I cannot do that." The reason for the HAL's malfunction and the movie's climax is a conflict in logic, a defect in the software.
Why it matters: According to MIT, there are over 100 errors in each thousand lines of code.
11 known deaths from human programming errors
A recent study reveals that software defects cost over $2 trillion per year
As reliance on AI grows, so do the stakes for software development
Between the lines: LLMs are not a solution for better software because they learn from the defects of human code.
Instead, we need new ways to approach writing software.
AI can help, but it is not the only solution.
What they're saying: According to VentureBeat, developers spend 20% of their time fixing bugs.
Computer scientist Jeff Atwood famously tweeted, "There are two hard things in computer science: cache invalidation, naming things, and off-by-one errors." While clever, it illustrates that humans cannot instruct computers without mistakes.
The solution: It starts with a conversation that leads to multiple technologies and methodology improvements so we do not turn 2001: A Space Odyssey from fiction to reality.
This newsletter was created as a line of communication between engineers, data scientists, and business leaders within AI to work together for safer code that results in safer AI. Your support keeps the conversation going.