The risks related to artificial intelligence (AI) are increasingly evident in a rapidly evolving technological context. Chancellor Rishi Sunak is under increasing pressure as academics and policy makers call for immediate regulation to address these concerns.
UK MPs have warned that any government regulation should focus on the potential harm AI poses to human life.
Concerns about public safety and national security were among a dozen issues raised by members of the Science, Innovation and Technology Committee as issues ministers must resolve before the UK hosts the first global conference at Bletchley Park. .
jump to
- Parliamentarians sound the alarm on AI: a deep dive into 12 issues that need to be addressed
- The Committee listed 12 challenges
- The NHS could benefit from AI in an exciting way
- The Government wants a “proportionate” approach
Parliamentarians sound the alarm on AI: a deep dive into 12 issues that need to be addressed
Credit: AP
At the event, held in November at Britain’s Second World War code-breaking base, Rishi Sunak and other leaders will examine the opportunities and risks posed by AI.
Alan Turing and others used Colossus computers to decipher data transmitted between Nazis on the ground, which was instrumental in the development of the technology.
The chairman of the committee, Conservative MP Greg Clark, said he “welcomed” the meeting but warned that the government may need to act with “greater urgency” to ensure future legislation does not become obsolete soon as nations like The United States, China and the EU explore their own guidelines on AI.
The Committee listed 12 challenges
1. Existential danger: If, as some experts have warned, AI poses a significant threat to human life, then regulation must include safeguards for national security.
2. Bias: AI has the potential to generate new prejudices or perpetuate current ones in society.
3. Confidentiality: Sensitive information about individuals or corporations could be used to train AI models.
4. Misrepresentation: Language models like ChatGPT can generate content that misrepresents someone’s actions, personal opinions, and character.
5. Data: the enormous amount of information needed to train the most powerful AI
6. Processing Power: Developing the most powerful AI requires massive processing power.
7. Transparency: AI models often need to explain why they create a specific result or where the data comes from.
8. Copyright: Generative models, whether text, images, music or videos, often use existing content, which must be protected so that the creative industries are not harmed.
9. Liability: If AI tools are exploited to cause harm, the policy must determine who is responsible, whether it is the developers or the vendors.
10. Employment: Politicians must anticipate the inevitable impact of AI adoption on existing jobs.
11. Openness and innovation: The computer code underlying AI models could be made publicly available to enable more reliable regulation and promote openness and innovation.
12. International coordination: Any standards must be developed internationally, and the November summit must include “as wide a range of countries as possible.”
Credit: Canva
The NHS could benefit from AI in exciting ways
Clark also mentioned healthcare as one of the “most interesting” areas of AI. It is already used in the NHS to analyze x-rays and scans, and academics are investigating how it could be used to predict dangerous long-term diseases such as diabetes.
He said AI could help make treatment “increasingly personalized, but highlighted the report’s concerns about potential biases being built into the training data of any AI model.
“If medical research is done on a specific sample or ethnic minority, the data that the AI is trained on may result in inaccurate recommendations,” he said.
Credit: Getty Images
The Government wants a “proportionate” approach
The committee stated that it would “in due course” issue a full set of recommendations to the government. He wants any AI bills to go before MPs during the next parliament, which will meet in September after the summer holidays.
A spokeswoman said the government is committed to a “proportionate and adaptable regulatory approach”, pointing to a £100m fund for the safe development of AI models in the UK.
“AI has enormous potential to change every aspect of our lives,” they added, “and we owe it to our children and grandchildren to harness that potential safely and responsibly.”
what do you think about it? Let us know in the comments.
For more trending stories, follow us on Telegram.
Categories: Trending
Source: vtt.edu.vn