[dead]
Took themselves off floppy disks and now look what happens. /s
[flagged]
The government's AI tool company likely had no understanding of context window limitations (although could be Iran or a false flag to be blamed on Iran) https://www.flyingmag.com/faa-smart-ai-now-active-at-22-artc...
Company that created the tool (called Airspace Intelligence) is interesting in terms of demographics for these types of companies - implies lots if the tool 'does not work as designed' (which it will obviously not)
https://www.linkedin.com/company/airspace-intelligence/peopl...
Maybe related to this? https://gizmodo.com/the-faa-is-rolling-out-its-ai-air-traffi...
Can't help but suspect that "AI" is to blame and the response is to sort of ban open source AI, or owning too much compute!
I wonder if it was Claude or ChatGPT? My money would be on the latter. Anthropic seems to generally have better security and guardrails, and apparently OpenAI’s newest cyber model is extremely capable.
No longer exists? Temporary glitch? (Edit: OP url was wrong https://www.fly.faa.gov/ois/oisedit/adv_oisedit_dis?number=1...)
...And Claude, remember NOT to make NO mistakes.
Let's just use GPU's to play Dark Souls like god intended and stop this nonsense.Yes, of course, who wouldn't just believe random statements from the worst prepared admin in history of America.
[flagged]
[flagged]
[flagged]