Subscribe
Sign in
General
p/generalShare and discuss tech, products, business, startups, or product recommendations
Start new thread
Stephen

Elon Musk and 1000 others call for a pause to advanced AI dev. What do you think?

by•

Website futureoflife.org posted an open letter and has collected over 1,000 signatures calling on all AI labs "to immediately pause for at least 6 months the training of AI systems more powerful than GPT-4." They say it is necessary to use the pause to assess the risks to society. I would love to know your thoughts on this.

Add a comment

Replies

Best
Brianna Swartz
The "slow AI" movement didn't start with Elon Musk and he shouldn't get the headline credit for it, in my opinion. I find what Mia Shah-Dand had this to say on this insightful: "Clarification: An open letter by high-profile tech folks calling for a moratorium on AI is trending on social media. This post is *not* an endorsement of that letter nor is it related to the efforts that Timnit Gebru Alex Hanna, Ph.D. and others have been leading since before 2021 with the launch of The Distributed AI Research Institute (DAIR). This is a reminder that women, Black scholars are the pioneers of the "Slow AI" movement not the powerful wealthy men who signed on to this letter. https://lnkd.in/gmbS-imJ #AIEthics #ResponsibleAI" https://spectrum.ieee.org/timnit...
Stephen
@brianna_swartz great point! While it's important to recognize the contributions of all those who are working to promote ethical and responsible AI, we should also strive to give credit where credit is due and amplify the voices of those who have been leading the charge.
Cher Williams
Are there risks to society? Absolutely. But risks currently exist for Elon's "Full Self Driving" Tesla technology - which isn't full self-driving at all and must be supervised at all times. I've also heard recent reports of numerous Tesla recalls, some of which are attributed to their steering wheels falling off - a problem that could effect 120K vehicles. THAT is a risk to me. Overall, I say follow the money. There is absolutely a monetary reason why Elon is pushing for this. I'd like to see other regulations in place to ensure no one person becomes even more of a billionaire as a result.
Stephen
@cher_williams While there are certain risks associated with the "Full Self Driving" Tesla technology, it's important to recognize that many technologies carry inherent risks. However, the specific issues mentioned, such as the steering wheel recalls, do highlight the need for careful regulation and oversight in the development and deployment of these technologies. Ultimately, it's important to balance the potential benefits of new technologies with the need to mitigate risks and ensure that they serve the public good, rather than simply enriching a few individuals.
Kevin Lu
I understand the concern and intention, but it's too naive to call a pause, unless Gov. stops it with legislation force. The pandora box has been opened, it's unstoppable!
Stephen
@dot_brand valid point! Despite the widespread adoption of the tech, it is still possible to take steps to mitigate their negative impacts. These include enhancing transparency and accountability, developing more responsible AI models, and prioritizing ethical considerations. While completely halting their development may not be feasible, it is important to prevent negative impacts and align their deployment with human values and ethics through research, collaboration, and responsible deployment.
Kevin Lu
@stephen_smith67 Truly! Slowing down the pace and inspecting the potential problems are important. However, the interest is so big that I believe there are too many people don't want to slow down, unless the Gov. forces to slow down the trend~
Valorie Jones
While large language models may have both positive effects (increased productivity) and negative effects (climate effects, proliferation of unsourced media) and wide-ranging economic impacts, the letter doesn't really propose any concrete actions on how to address any of these concerns, short of an indeterminate pause. Also, bigger models exclusively mean better models. Part of solution will be new models that have more validation and safety checks built in, and to find new ways more efficient methods of training and evaluating these models.
Stephen
@val_jones Indeed, the concerns surrounding large language models are complex and multi-faceted, and there is no single silver bullet solution. However, it is clear that a proactive approach is needed to address the negative impacts while still leveraging the benefits of these models. This will require a combination of measures, such as developing new models that prioritize validation and safety, implementing more efficient training and evaluation methods, and fostering greater transparency and accountability in the use of these models. Ultimately, it will take a collaborative effort from researchers, policymakers, and industry leaders to strike a balance between innovation and responsibility in the development and deployment of large language models.