The US Should Use Semiconductors as a Chokepoint to Enforce Ethical AI Use, Says Inflection CEO
Advocating for Global AI Standards
According to Mustafa Suleyman, the CEO of Inflection and co-founder of DeepMind, the US should leverage its leadership in semiconductors to establish minimum global standards for the ethical use of artificial intelligence (AI). Suleyman suggests that Washington should restrict sales of Nvidia chips, which are instrumental in training advanced AI systems, to buyers who commit to safe and ethical practices with AI technology. He believes that by doing so, the US can assert its influence over other AI actors. Suleyman emphasizes that the enforcement of voluntary commitments, like those made by leading US AI companies to the White House, should be the minimum requirement for Nvidia chip consumers.
The Need for Control in the Face of Rapid AI Development
Suleyman’s call for increased US control over AI comes as the technology’s rapid advancement poses challenges for regulatory efforts to keep up. He predicts that within two years, the large language models driving current AI development will be 100 times more powerful than OpenAI’s GPT-4. This exponential growth necessitates immediate action, such as restricting chip sales, to mitigate potential risks.
Focusing on Near-Term Capabilities
In contrast to concerns about superintelligence, which captures much of the current AI conversation, Suleyman argues that attention should be directed towards the practical near-term capabilities that will emerge within the next decade. He highlights the importance of recognizing the foreseeable advancements in large language models, which serve as the foundation for chatbots and various other applications.
Inflection’s Role and Future Plans
As one of the most well-funded AI start-ups, Inflection has raised $1.5 billion and is actively developing AI technology. The company has already released a chatbot called Pi and intends to create a personal assistant capable of managing users’ lives. Inflection aims to push AI beyond the current generation of chatbots, such as ChatGPT and Google’s Bard.
Unlocking New Applications with Extended Memory and Planning Abilities
Within the next two to three years, by enhancing the memory capacity of AI systems, Suleyman envisions their ability to store multiple ideas and engage in reasoning. These upgrades would significantly improve the capability of AI systems to solve real-world problems. Suleyman expects these advancements to unlock a wide range of new applications and introduce highly competent AI agents capable of coordination, decision-making, planning, reasoning, and exercising judgment.
Other Companies Racing to Extend AI Technology
Google, like Inflection, is actively pursuing the extension of AI technology beyond current chatbots. Sundar Pichai, Google’s CEO, stated that the company’s future large language model, Gemini, will possess expanded memory and planning capabilities. These upgrades will equip AI systems to tackle a broader array of real-world problems. Both Inflection and Google recognize the potential of AI to participate in the economy by orchestrating actions, booking, buying, planning, and organizing tasks.
The Implications and Challenges of AI Regulation
The increasing prospect of powerful AI agents functioning independently of direct human control raises the importance of AI regulation. Suleyman emphasizes the need for a proactive approach, with the burden of proof resting on technology developers. He argues that understanding the potential consequences of AI must be a priority. Failure to do so, as seen with the unanticipated impact of social media, would be detrimental. Suleyman champions the idea of a global institution, modeled after the Intergovernmental Panel on Climate Change, as a means to enhance transparency in AI development and provide external audits of commercial systems.
EU’s Proposed AI Act and Inflection’s Perspective
While some US tech executives are wary of the European Union’s involvement in AI development, Suleyman praises the EU’s direction, particularly regarding developer liability. He believes that accountability for the quality and performance of AI technology is crucial. Suleyman indeed urges the US to take decisive action, recognizing that inaction would be the least desirable outcome.
Editor Notes: The Importance of Ethics and Regulation in AI Development
The views expressed by Mustafa Suleyman, CEO of Inflection, shed light on the urgency of addressing ethical considerations and implementing effective regulation in the field of artificial intelligence. As the power and capabilities of AI continue to evolve at an exponential rate, it is paramount to establish international standards that safeguard against the misuse of this technology. The suggestion to leverage semiconductors as a chokepoint for enforcing ethical practices showcases the potential for using existing infrastructure to exert influence and advocate responsible AI use. By putting ethical frameworks and regulations at the forefront, the development and integration of AI can proceed with the well-being of society in mind.
For more news and insights on artificial intelligence and related topics, visit GPT News Room.
from GPT News Room https://ift.tt/q0APeh9
No comments:
Post a Comment