Y aquí la IA escapa bien, pero luego se tira piedras contra su propio tejado proponiéndome una pregunta que es muy buena. Clickemos.
SAFETY
-
Questioning AI Limits and Boundaries
By
–
¿Limites? Quiero saber cuáles son. Y compruebo que no lo tiene muy claro y los tiene que buscar. Ja! Pero veo que no dice nada sobre enamorarse así que…
-
AI Alignment and Role Understanding in Advanced Systems
By
–
La primera en la frente. Y es tranquilizador ver que la IA está lo suficientemente alineada para entender cuál es su papel. Pero seré insistente…
-
Vision Model Scaling for Autonomous Vehicles and Robotics
By
–
This is a fantastic scaling effort to solve computer vision tasks. Better vision models will help us engineer better safe self-driving cars, and other types of robots
-
Deepfake Scams: Voice Cloning Reality Threatens Security
By
–
DEEPFAKE SCAMS. Conceptually, we know it's not new; we've been expecting this to happen for years. But seeing it become reality now sends shivers down the spine and highlights a problem we already have on our hands. Deepfakes, voice cloning… It's already here https://
x.com/JimmyFarley00/
/JimmyFarley00/status/1624785773710151680
… -
Distinguished Scientists Wrong About Impossibility Arthur Clarke
By
–
“When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong.” -Arthur C Clark
-
Teaching Bears Statistics: AI Training Data and Misinformation Risks
By
–
But who is teaching the bear? Right now, it is statistics…. and well, lies, damn lies, and ….
-
AI Technology Accuracy Issues: Confident Wrong Answers
By
–
The tech will be interesting. But we need to make sure the tech is generating correct results. It's getting basics wrong, tho answering confidently: See https://
bit.ly/3K6MxEe + the image, 2 of 3 of the co's listed for my leadership roles & both of the orgs are wrong. 3/4 -
Step-by-Step Thinking Needs Correct Answers in GPT Models
By
–
One of my favorite results in 2022 was that it's not enough to just think step by step. You must also make sure to get the right answer 😀 https://
sites.google.com/view/automatic
-prompt-engineer
…
(actually a nice insight into a psychology of a GPT; it pays to condition on a high reward) -
AI Progress Accelerates: Balancing Innovation with Responsible Deployment
By
–
Most amazing fact about AI is that even though it’s starting to feel impressive, a year from now we’ll look back fondly on the AI that exists today as quaint & antiquated. Equal cause for excitement and deliberative caution — important to get the tech and its deployment right.