Digital Ethics

1- Fireside Chat of Hon. Minister with Industry Leaders and invited Ministers (Public) 10:00 AM – 11:00 AM2- Mitigating Social Risks for a Safe and Resilient...

Researchers worry AI bots like Grok are already showing signs of larger-scale problems.

What does life online look like filtered through a bot?

In the last year, chatbots powered by large language models are everywhere and even … useful. But how do they work?

A group of Stanford researchers found that large language models can propagate false race-based medical information.

The rise of ChatGPT in schools led to an influx of unproven AI detectors. But results have shown this approach may have more consequences than benefits.

Some senior employees described Sam Altman as psychologically abusive, creating chaos at OpenAI. The complaints were a major factor in the board’s decision to fire the CEO.

En veileder for å avdekke og forebygge diskriminering i utvikling og bruk av kunstig intelligens

So, about this "truly anonymous" synthetic data… 😬 I've been telling anyone who'd listen that ad hoc approaches to anonymization should be assumed broken…

Training generative models to produce synthetic data is meant to provide a privacy-friendly approach to data release. However, we get robust guarantees only when models are trained to satisfy...

Observer investigation reveals Meta Pixel tool passed on private details of web browsing for targeted advertising purposes

Visit the post for more.

A new study has found that the use of AI tools such as ChatGPT in the workplace entrenches biased language based on gender

A new study has found that the use of AI tools such as ChatGPT in the workplace entrenches biased language based on gender

Researchers were able to purchase all kinds of information about military members, from their net worth to their homeowner status. That’s a national security disaster.

This is the first time the carbon emissions caused by using an AI model for different tasks have been calculated.

ContentsExecutive SummaryBreaking it Down: What Does Comprehensive Data Privacy Legislation Look Like?Sketching the Landscape: What Real Privacy Protections Might Accomplish Protecting Children’s Mental Health Supporting Journalism Protecting Access to Healthcare Fostering Digital Justice...

A whistleblower believes the self-driving vehicle technology is not safe enough for public roads.

By centering peoples' rights we can shape AI to maximally benefit humanity while minimizing harms, says Margaret Mitchell.

AI can be harnessed for social good, such as a new app to help mothers navigate paid leave policies, writes Reshma Saujani of Girls Who Code.

Generative AI: The Data Protection Implications (CEDPO). How data is used to train generative AI systems. This takes place in four broad ways: 1. Based on…

The country has come to rely on Clearview AI for a range of wartime tasks, many of which have not been reported:

One year after the launch of ChatGPT, a question lingers: Is there an ethical way to use AI writing? If so, what steps does one need to take?

Supposed AI breakthroughs are frequently veiled in secrecy, hindering scientific consensus.

Fiber-optic cables that were damaged by a rockfall in an undersea canyon, resulting in slow internet connections in some parts of Africa, should be repaired next month by a specialized vessel, according to telecommunication companies.

'AI thrives when our need for originality is low and our demand for mediocrity is high,' writes Ray Nayler.

Matthew Butterick is leading a wave of lawsuits against major AI firms, from OpenAI to Meta. Win or lose, his work will shape the future of human creativity.

Llama 2-Chat is a collection of large language models that Meta developed and released to the public. While Meta fine-tuned Llama 2-Chat to refuse to output harmful content, we hypothesize that public access to model weights enables bad actors to cheaply circumvent Llama 2-Chat's safeguards and weaponize Llama 2's capabilities for malicious purposes. We demonstrate that it is possible to effectively undo the safety fine-tuning from Llama 2-Chat 13B with less than $200, while retaining its general capabilities. Our results demonstrate that safety-fine tuning is ineffective at preventing misuse when model weights are released publicly. Given that future models will likely have much greater ability to cause harm at scale, it is essential that AI developers address threats from fine-tuning when considering whether to publicly release their model weights.

2023 LLVM Developers' Meetinghttps://llvm.org/devmtg/2023-10------A Technical Guide to Compassion in Open Source Speaker: April Wensel------Slides: https://...