In our ongoing ChatGPT series, which focuses on the potentialities of generative AI - such as ChatGPT -, and how it impacts our rights and our everyday reality, we inevitably also discuss the risks and needs - or otherwise - for regulation, in its various possible forms.
In this regard, we feel it appropriate to repost the open letter on 'Pause Giant AI Experiments' - also purely to share comprehensive information on the topic. The letter, signed by world experts in the field (including Yoshua Bengio, Stuart Russell and Steve Wozniak and many other relevant names), depicts how AI systems with human-like intelligence pose significant risks to society and require careful planning and management. It is argued as AI labs are engaged in a race to develop and deploy ever more powerful digital minds without sufficient control or understanding. As AI systems become competitive with humans, the initiative calls for consideration of the potential impact on society and humanity, including the automation of jobs and the development of non-human minds that could overtake and replace us. Indeed, based on the latter consideration, the development of powerful AI systems should only proceed when we are sure that their effects will be positive and their risks manageable.
The Future of Life Institute and the signatoires call on AI labs to immediately suspend training of AI systems more powerful than GPT-4 for at least six months and use this time to develop shared safety protocols for the design and development of advanced AI. It is stated as AI research and development should focus on making current systems more accurate, secure, transparent and reliable, while policymakers should accelerate the development of robust AI governance systems that include oversight, auditing and certification. Let us enjoy the long summer of AI, let us not rush unprepared into the autumn.
For more information, see the letter and petitioners here: https://futureoflife.org/open-letter/pause-giant-ai-experiments/
If you are undecided about whether or not to sign the letter, here is ChatGPT's response:
"The decision to sign or not sign the "Pause Giant AI Experiments: An Open Letter" is a matter of personal judgement and perspective. However, I can provide some reasons that have been presented by both sides of the debate:
Reasons to sign the letter:
- Concerns over the potential societal implications of large language models, such as exacerbating economic inequality, perpetuating racial biases, and threatening privacy and democracy.
- A need for more investment in research on the societal implications of large language models.
- A call for transparency and interpretability in AI systems, so that people can understand how they work and what they are being used for.
- A belief that the potential risks and consequences of large language models should be fully understood and addressed before further development takes place.
- A desire to prioritize ethical considerations in AI development and use.
Reasons not to sign the letter:
- A belief that large language models have the potential to bring significant benefits, such as improved natural language processing, language translation, and other applications.
- A concern that pausing research and development of large language models could slow down progress in the field of AI and limit its potential to solve important problems.
- A belief that AI development can be guided by ethical principles and regulations, and that the potential risks of large language models can be mitigated through responsible development and use.
- A desire to continue advancing AI technology while also addressing ethical and societal concerns".
Luca Megale
is a PhD Student at LUMSA University of Rome
and tutor of the European Master in Law and Economics - EMLE (Rome term)