The boring checklist still wins:
- never give the model direct access to secrets
- scope tool permissions tightly
- validate outputs before using them
- log prompts, tool calls, and weird edge cases
AI security is mostly software security wearing a new hoodie.
Same bug. New interface.
Prompt injection isn’t weird AI magic.
It’s the oldest security lesson with better branding: don’t trust user-controlled input.
Treat LLM output as untrusted data. Sanitize it. Limit permissions. Log the weird stuff.
#AIsecurity#LLMSecurity
Two years of nights and weekends on this project.
One weekend with AI.
It moved forward more in that weekend than in the previous 2 years combined.
That’s not hype. That’s a workflow shift.
https://t.co/aD7bH4Djd2
#BuildInPublic#AI#Gamedev#IndieDev#CRYENGINE
Peak AI behavior: I watched a video about the risks of AI-assisted coding… https://t.co/00nmaE0REN
then used AI to help me research the effects of using AI for daily coding tasks 😂
Turned the findings into this infographic.
Is AI making us better developers, or just faster at trusting bad answers? Debate below 👇
🚨 Voyager 1 will be one light-day from Earth on Nov. 13, 2026. A signal from the probe—traveling at light speed (≈300,000 km/s)—will take about 24 hours to reach us as it continues its trek through interstellar space.
Thanks to everyone 🩵 for trusting us at @EFDevcon 🔥
120 more bookings.
22 more countries 🌍
All paid in crypto.
All fully secure.
🇩🇪🇦🇷🇪🇸🇺🇸🇮🇹🇮🇳🇲🇽🇹🇷 — and more.
#TrustFirst#DevConnect