USA: There has been an increase in the volume of calls for AI regulation. It's interesting to note that OpenAI, the organisation that ignited the current AI frenzy, is leading those calls. Recently, the start-up voiced its opinions regarding the introduction of a democratic process to control AI systems. It has now started a grant programme to support such a system.
The term "generative AI" is in vogue everywhere. The general public is interested in technology that can simulate human intelligence. It has also raised concerns about the possible harm it might do. Everyone is interested in learning how businesses and lawmakers intend to control AI tools. Given its prominent position in the AI industry, OpenAI's opinion in this debate is valuable.
Also Read: Watch! American Lawmakers to unveil Taiwan war to-do list
A $1 million programme is being launched by OpenAI to support research into creating a "democratic process for deciding what rules AI systems should follow." Both individuals and organisations are eligible for the programme. The startup wants to identify 10 qualified candidates. Each applicant who is chosen will be given a grant of $100,000 to test out their idea as it is presented in the application.
The co-founders of OpenAI, Sam Altman, Greg Brockman, and Ilya Sutskever, discussed the necessity of regulating "superintelligence" or advanced AIs in a blog post earlier this week.
People should "democratically decide" the "bounds and defaults" of AI systems, they claimed. However, they did not specify how the guardrails would be used. According to what the company said, the grant appears to be the next step.
What exactly does 'democratic process' mean in the context of OpenAI? In the blog post about the grant, the company also explains what it means by a "democratic process."
A "broadly representative group of people exchange opinions, engage in deliberative discussions, and ultimately decide on an outcome via a transparent decision-making process," according to OpenAI, is involved in this process. Doesn't that sound like the perfect definition of democracy?
Also Read: AI-powered drone swarms are tested by the anti-China bloc
Candidates can choose from a list of questions provided by OpenAI to demonstrate their strategy. Some of these concerns are "How should AI assistants respond to questions about public figure viewpoints?" and "Under what circumstances, if any, should AI assistants be permitted to offer medical, financial, or legal advice?" Interested parties have until June 24 to respond to one or more of the listed questions.