When people talk about issues with social networks, they don't realize that AI is not the cause, AI is actually an essential piece of the solution. To take down propaganda, hate speech, attacks on democracy, child exploitation, calls to violence, dangerous misinformation, etc in
ETHICS
-
Advisory Board Composition Raises Corporate Interest Concerns
By
–
That's great news — reading through the list of advisors, I'm pleasantly surprised to discover it's not entirely stacked with corporate special interests. (Although there are still quite a large bunch of them…)
-
7 Tactics to Reduce Hallucinations in Large Language Models
By
–
In this one: 7 tactics to reduce hallucinations in large language models (LLMs) in this video. Tactics include adjusting inference parameters, improving prompt engineering, and other techniques to bolster LLM's reliability and accuracy. Let's foster trustworthy AI today!
-
Dynamic Memory LLMs Will Obsolete Prompt Engineering
By
–
Now we have LLMs with a fixed-size context. Imagine LLMs with dynamic, expandable long-term memory. You'll align them to yourself through iterative conversations. This will render prompt engineering obsolete. The LLM will anticipate your needs. The challenge will be: if the
-
Exceptional Prompts as Trade Secrets in AI Era
By
–
We are entering an era where certain prompts will be treated as trade secrets because they are that exceptional: – hard to discover
– doing exactly what's intended
– avoiding unintended outcomes
– being concise and efficient They represent perfect alignment between the -
MLCommons develops standard AI safety benchmarks for researchers
By
–
Today on the blog, learn how we’re supporting a new effort by the non-profit MLCommons Association that aims to bring together expert researchers across academia and industry to develop standard AI safety benchmarks that everyone can use and understand. ↓
-
New Quantitative Evidence-Driven Approach to AI Safety
By
–
We are working on a a new quantitative & evidence-driven approach to AI safety, beyond what is commonly thought to be possible (and we’re hiring!):
-
Supreme Court faces AI liability questions, scholars urge action
By
–
The Supreme Court will soon face questions of liability for harmful outputs from generative AI. Stanford legal scholars urge policymakers to get ahead of the storm.
-
New AI Preparedness Team Evaluates AGI Risks Quantitatively
By
–
We are building a new Preparedness team to evaluate, forecast, and protect against the risks of highly-capable AI—from today's models to AGI. Goal: a quantitative, evidence-based methodology, beyond what is accepted as possible:
-
Commercial pressures shape algorithm design choices
By
–
“Sort by engagement” is a choice to make your life worse and I wish we did not feel such strong commercial pressures to default users to it.