Why did OpenAI use Scarlett Johansson's voice? As Jessa Lingel & I discuss in our journal article on AI agents, there's a long history of using white women's voices to “personalize” a technology to make it feel safe and non-threatening while it is capturing maximum data.
SAFETY
-
Scaling Monosemanticity: Understanding Neural Network Features and Safety
By
–
There’s much more in our paper, including detailed analysis of the breadth and specifics of features, many more safety-relevant case studies, and preliminary work on using features to study computational "circuits" in models. Read the full paper here: https://
transformer-circuits.pub/2024/scaling-m
onosemanticity/index.html
… -
Safety Features Evaluation Requires Further Practical Validation
By
–
This work is preliminary. Whereas we show that there are many features that seem *plausibly* relevant to safety applications, much more work is needed to establish that our approach is useful in practice.
-
Claude’s Secrecy Feature: Information Withholding Mechanism
By
–
One notable example is a "secrecy" feature. We observe that it fires for descriptions of people or characters keeping a secret. Activating this feature results in Claude withholding information from the user when it otherwise would not.
-
Model Safety Features: Vulnerabilities, Deception, and Bias Detection
By
–
Among these millions of features, we find several that are relevant to questions of model safety and reliability. These include features related to code vulnerabilities, deception, bias, sycophancy, power-seeking, and criminal activity.
-
Dictionary Learning Makes LLM Neurons More Interpretable
By
–
The problem: most LLM neurons are uninterpretable, stopping us from mechanistically understanding the models. In October, we showed that dictionary learning could decompose a small model into "monosemantic" components we call "features"—making the model more interpretable.
-
Concerns About Hidden Practices and Lack of Transparency
By
–
I get such dodgy feelings about them trying to hide WHAT they're doing so bad
-
GPT-4o Safety Concerns: OpenAI’s Approach to Unsafe Queries
By
–
It certainly makes you wonder what it is that OpenAI has in GPT-4o that allows that service to (presumably) answer the query without safety concerns (unless in fact it *is* unsafe and OpenAI has done nothing about it.)
-
Software Reliability at Scale: Testing and Safety Nets Matter
By
–
“A lot of the traditional ways of thinking of software at scale haven't changed. The way you have to approach reliability, the way that you get trust through testing, the fact that you do really need those safety nets.
— DataRobot (@DataRobot) 21 mai 2024
Safety nets might even be more important than the… pic.twitter.com/mpK3u4AIDp“A lot of the traditional ways of thinking of software at scale haven't changed. The way you have to approach reliability, the way that you get trust through testing, the fact that you do really need those safety nets. Safety nets might even be more important than the
-
OpenAI voice controversy: Scarlett Johansson vs ChatGPT similarity debate
By
–
El tema del día es la voz de Scarlett Johanson vs la voz de ChatGPT. Y tal y como recoge Carlos en este hilo las voces realmente no son idénticas.
— Carlos Santana (@DotCSV) 21 mai 2024
Que OpenAI haya querido usar la voz de Scarlett para aproximar el efecto Her no significa que finalmente lo hayan hecho. ¿Opiniones? https://t.co/pD9LVGoaRAEl tema del día es la voz de Scarlett Johanson vs la voz de ChatGPT. Y tal y como recoge Carlos en este hilo las voces realmente no son idénticas. Que OpenAI haya querido usar la voz de Scarlett para aproximar el efecto Her no significa que finalmente lo hayan hecho. ¿Opiniones?