Mira Murati fears that AI can be misused and "used by bad actors."
ChatGPT is a popular artificial intelligence (AI) chatbot which is capable of generating human-like text based on the input it is given. The revolutionary technology, launched in November 2022, has rapidly become the internet's favourite tool as it can answer complex questions conversationally. However, like every coin has two sides, there are also many dangerous things that the ChatGPT can do and experts are increasingly concerned about its extraordinary capabilities. From cybersecurity threats to fears that it might make many jobs obsolete, many people are sceptical about its potential implications.
Recently, Mira Murati, CTO at OpenAI, the company that developed ChatGPT and Dall-E, expressed her concerns about the chatbot in an interview with Time magazine. Ms Murati fears that AI can be misused and "used by bad actors."
She told Time magazine, ''[AI] can be misused, or it can be used by bad actors. So, then there are questions about how you govern the use of this technology globally. How do you govern the use of AI in a way that's aligned with human values?''
She added that it's not too late for different stakeholders to get involved and some regulations may be needed.
When asked whether companies like OpenAI or governments should be in charge of regulating the tool, MS Murati said, "It's important for OpenAI and companies like ours to bring this into the public consciousness in a way that's controlled and responsible.''
She further stressed that the company will need all the help it can get, including from regulators, governments, and everyone else. ''But we're a small group of people and we need a ton more input in this system and a lot more input that goes beyond the technologies--definitely regulators and governments and everyone else.''
Notably, ChatGPT was created by San Francisco-based artificial intelligence company OpenAI. OpenAI is famous for its well-known DALL-E, a deep-learning model that generates images from text instructions called prompts.
Meanwhile, OpenAI has also acknowledged that ChatGPT can give incorrect answers on multiple occasions, and the official website notes that the chatbot might even "produce harmful instructions or biased content."