There's a chance that ChatGPT knows personal details about you--and if it doesn't, it might just make something up. As OpenAI's generative text chatbot has boomed in popularity over the past six months, the risks of the system being trained on data vacuumed up from the web have become clearer. Data regulators around the world are investigating issues with how OpenAI gathered the data it uses to train its large language models, the accuracy of answers it provides about people, and other legal concerns about the use of its generative text systems. Europe's data regulators have joined forces to look at OpenAI after Italy temporarily banned ChatGPT from the country. And Canada is also investigating the technology's potential privacy risks.

In AI development, the dominant paradigm is that the more training data, the better. OpenAI's GPT-2 model had a data set consisting of 40 gigabytes of text. GPT-3, which ChatGPT is based on, was trained on 570 GB of data. OpenAI has not shared how big the data set for its latest model, GPT-4, is. But that hunger for larger models is now coming back to bite the company. In the past few weeks, several Western data protection authorities have started investigations into how OpenAI collects and processes the data powering ChatGPT.

AI developers have "no excuse" for getting data privacy wrong, one of the heads of the UK's data regulator has said, warning those who don't follow the law on data protection will face consequences. The Information Commissioner's Office (ICO) enforces data protection in the UK. Speaking amid the explosion of interest in generative AI, especially Large Language Models like the one that powers OpenAI's ChatGPT, Stephen Almond, the ICO's executive director of regulatory risk, warned LLMs posed a risk for data security. Writing in a blog post, he argued it is time to "take a step back and reflect on how personal data is being used". He noted that Sam Altman, the CEO of ChatGPT creator OpenAI, has himself declared his own worries about AI advances and what they could mean.

The UK's data regulator has issued a warning to tech companies about protecting personal information when developing and deploying large language, generative AI models. Less than a week after Italy's data privacy regulator banned ChatGPT over alleged privacy violations, the Information Commission's Office (ICO) published a blog post reminding organizations that data protection laws still apply when the personal information being processed comes from publicly accessible sources. "Organisations developing or using generative AI should be considering their data protection obligations from the outset, taking a data protection by design and by default approach," said Stephen Almond, the ICO's director of technology and innovation, in the post. Almond also said that, for organizations processing personal data for the purpose of developing generative AI, there are various questions they should ask themselves, centering on: what their lawful basis for processing personal data is; how they can mitigate security risks; and how they will respond to individual rights requests. "There really can be no excuse for getting the privacy implications of generative AI wrong," Almond said, adding that ChatGPT itself recently told him that "generative AI, like any other technology, has the potential to pose risks to data privacy if not used responsibly."

When OpenAI released GPT-3 in July 2020, it offered a glimpse of the data used to train the large language model. Millions of pages scraped from the web, Reddit posts, books, and more are used to create the generative text system, according to a technical paper. Scooped up in this data is some of the personal information you share about yourself online. This data is now getting OpenAI into trouble. On March 31, Italy's data regulator issued a temporary emergency decision demanding OpenAI stop using the personal information of millions of Italians that's included in its training data.

Companies of all kinds use machine learning to analyze people's desires, dislikes, or faces. Some researchers are now asking a different question: How can we make machines forget? A nascent area of computer science dubbed machine unlearning seeks ways to induce selective amnesia in artificial intelligence software. The goal is to remove all trace of a particular person or data point from a machine learning system, without affecting its performance. If made practical, the concept could give people more control over their data and the value derived from it.

Companies of all kinds use machine learning to analyze people's desires, dislikes, or faces. Some researchers are now asking a different question: How can we make machines forget? A nascent area of computer science dubbed machine unlearning seeks ways to induce selective amnesia in artificial intelligence software. The goal is to remove all trace of a particular person or data point from a machine learning system, without affecting its performance. If made practical, the concept could give people more control over their data and the value derived from it.

Technical reliability: Smaller challenges within this broader area include deep neural networks (DNNs) lack of robustness resulting in susceptibility to invasion attacks. Lack of transparency and explainability of these complex systems may infringe legal or regulatory requirements such as GDPR in terms of automated decision-making. Also, data breaches, tampering, theft, and misuse of vast amounts of data may result from the unavailability of comprehensive data security protection. A case of this is in the field of autonomous driving when evasion attacks can lead to traffic offenses and even trigger accidents. In healthcare, attackers can also introduce significant errors in the dosage recommended by AI models by only adding a small amount of malicious data but resulting in huge serious consequences.

I think it's amazing that you can design algorithms that might have society-level impact. AI algorithms are in that category. Take Genetic Programming, where you write computer programs to evolve other computer programs. Moreover, AI poses exciting engineering challenges like scaling, and it asks the biggest questions, like the nature of the mind. For these reasons, AI was my first love.

I think it's amazing that you can design algorithms that might have society-level impact. AI algorithms are in that category. Take Genetic Programming, where you write computer programs to evolve other computer programs. Moreover, AI poses exciting engineering challenges like scaling, and it asks the biggest questions, like the nature of the mind. For these reasons, AI was my first love.