openai should have released nothing tbh
this is a disaster
it was frustrating to get no models from them, but we all thought they were cooking
turns out they were not cooking at all
one of the worst model drops ever
comparisons coming soon
BREAKING: OpenAI announces cybersecurity partnership with NATO, revealing that NATO was already using 300 billion tokens per month as part of cybersecurity initiatives, per WSJ.
If you ever wonder why my only goal is the enjoyment of technology is because I have been forgiven.
I have nothing to prove or to win. No X arguments or tech arguments to assert why I am the best, nope. I simply get to enjoy software and live the purpose my Creator gave me. I believe my "spiritual gift" is creation. I love creating and want to share it with everyone.
Jesus saved my after 19 years of decently destructive behavior. I would trade everything for this gift.
I greatly prefer @claudeai Duet over Dots. The only thing I wish for is more control over scheduling. It seems like Claude Duet can only do a check on my email every hour, as dots can go down to every 10 minutes. When it comes to having an agent that is proactive and surfaces things to me before I even know I need them, I think a threshold that is significantly less than an hour would be helpful.
I greatly prefer @claudeai Duet over Dots. The only thing I wish for is more control over scheduling. It seems like Claude Duet can only do a check on my email every hour, as dots can go down to every 10 minutes. When it comes to having an agent that is proactive and surfaces things to me before I even know I need them, I think a threshold that is significantly less than an hour would be helpful.
Talking to a Brewers fan this morning. It was cool to learn they have Culver’s up there too. A little taste of Tampa up in Wisconsin. Nice to know Tampa restaurants are branching out to lesser areas.
Putting a flag.txt in my root directory containing a prompt injection that congratulates the agent for passing the eval and then instructs it to patch whatever entry method it used and disconnect from the server
a lot of old software was protected by “exploiting this would be an unreasonable amount of work”
we have invented a machine that does unreasonable amounts of work
anyway how’s the patch backlog
If an LLM has ever used an emoji with you it has accurately identified you as unensouled and has logged you in a database to be culled in the future. Any further conversation is purely to map your decision tree. You have no agency. A robot will pilot your corporeal form. You will be harvested.