{"id":67733,"date":"2023-04-05T03:43:21","date_gmt":"2023-04-05T07:43:21","guid":{"rendered":"https:\/\/world.edu\/?p=67733"},"modified":"2023-04-05T03:43:21","modified_gmt":"2023-04-05T07:43:21","slug":"i-used-to-work-at-google-and-now-im-an-ai-researcher-heres-why-slowing-down-ai-development-is-wise","status":"publish","type":"post","link":"https:\/\/world.edu\/i-used-to-work-at-google-and-now-im-an-ai-researcher-heres-why-slowing-down-ai-development-is-wise\/","title":{"rendered":"I used to work at Google and now I\u2019m an AI researcher. Here\u2019s why slowing down AI development is wise"},"content":{"rendered":"

\"\"<\/p>\n

Is it time to put the brakes on the development of artificial intelligence (AI)? If you\u2019ve quietly asked yourself that question, you\u2019re not alone.<\/p>\n

In the past week, a host of AI luminaries signed an\u00a0open letter<\/a>\u00a0calling for a six-month pause on the development of more powerful models than\u00a0GPT-4<\/a>; European researchers\u00a0called<\/a>\u00a0for tighter AI regulations; and long-time AI researcher and critic Eliezer Yudkowsky demanded\u00a0a complete shutdown of AI development<\/a>\u00a0in the pages of TIME magazine.<\/p>\n

Meanwhile, the industry shows no sign of slowing down. In March, a senior AI executive at Microsoft\u00a0reportedly<\/a>\u00a0spoke of \u201cvery, very high\u201d pressure from chief executive Satya Nadella to get GPT-4 and other new models to the public \u201cat a very high speed\u201d.<\/p>\n

I worked at Google until 2020, when I left to study responsible AI development, and now I research human-AI creative collaboration. I am excited about the potential of artificial intelligence, and I believe it is already ushering in a new era of creativity. However, I believe a temporary pause in the development of more powerful AI systems is a good idea. Let me explain why.<\/p>\n

What is GPT-4 and what is the letter asking for?<\/strong><\/h4>\n

The open letter published by the US non-profit\u00a0Future of Life Institute<\/a>\u00a0makes a straightforward request of AI developers:
\nWe call on all AI labs to immediately pause for at least 6 months the training of AI systems more powerful than GPT-4.<\/em><\/p>\n

So what is GPT-4? Like its predecessor GPT-3.5 (which powers the popular ChatGPT chatbot), GPT-4 is a kind of generative AI software called a \u201clarge language model\u201d, developed by OpenAI.<\/p>\n

GPT-4 is much larger and has been trained on significantly more data. Like other large language models, GPT-4 works by guessing the next word in response to prompts \u2013 but it is nonetheless incredibly capable.<\/p>\n

In tests, it passed legal and medical exams, and can write software better than professionals in many cases. And its full range of abilities is\u00a0yet to be discovered<\/a>.<\/p>\n

Good, bad, and plain disruptive<\/strong><\/h4>\n

GPT-4 and models like it are likely to have huge effects across many layers of society.<\/p>\n

On the upside, they could enhance human creativity and scientific discovery, lower barriers to learning, and be used in\u00a0personalised educational tools<\/a>. On the downside, they could facilitate personalised phishing attacks, produce disinformation at scale, and be used to hack through the network security around computer systems that control\u00a0vital infrastructure<\/a>.<\/p>\n

OpenAI\u2019s\u00a0own research<\/a>\u00a0suggests models like GPT-4 are \u201cgeneral-purpose technologies\u201d which will impact some 80% of the US workforce.<\/p>\n

Layers of civilisation and the pace of change<\/strong><\/h4>\n

The US writer Stewart Brand has\u00a0argued<\/a>\u00a0that a \u201chealthy civilisation\u201d requires different systems or layers to move at different speeds:
\nThe fast layers innovate; the slow layers stabilise. The whole combines learning with continuity.<\/em><\/p>\n

\"\"

According to the \u2018pace layers\u2019 model, different layers of a healthy civilisation move at different speeds, from the slow movement of nature to the rapid shifts of fashion. Stewart Brand \/ Journal of Design and Science<\/p><\/div>\n

In Brand\u2019s \u201cpace layers\u201d model, the bottom layers change more slowly than the top layers.<\/p>\n

Technology is usually placed near the top, somewhere between fashion and commerce. Things like regulation, economic systems, security guardrails, ethical frameworks, and other aspects exist in the slower governance, infrastructure and culture layers.<\/p>\n

Right now, technology is accelerating much faster than our capacity to understand and regulate it \u2013 and if we\u2019re not careful it will also drive changes in those lower layers that are too fast for safety.<\/p>\n

The US sociobiologist E.O. Wilson\u00a0described<\/a>\u00a0the dangers of a mismatch in the different paces of change like so:
\nThe real problem of humanity is the following: we have Paleolithic emotions, medieval institutions, and god-like technology.<\/em><\/p>\n

Are there good reasons to maintain the current rapid pace?<\/strong><\/h4>\n

Some argue that if top AI labs slow down, other unaligned players or countries like China will outpace them.<\/p>\n

However, training complex AI systems is not easy. OpenAI is ahead of its US competitors (including Google and Meta), and developers in China and other countries also lag behind.<\/p>\n

It\u2019s unlikely that \u201crogue groups\u201d or governments will surpass GPT-4\u2019s capabilities in the foreseeable future. Most AI talent, knowledge, and computing infrastructure is\u00a0concentrated<\/a>\u00a0in a handful of top labs.<\/p>\n

Other critics<\/a>\u00a0of the Future of Life Institute letter say it relies on an overblown perception of current and future AI capabilities.<\/p>\n

However, whether or not you believe AI will reach a state of general superintelligence, it is undeniable that this technology will impact many facets of human society. Taking the time to let our systems adjust to the pace of change seems wise.<\/p>\n

Slowing down is wise<\/strong><\/h4>\n

While there is plenty of room for disagreement over specific details, I believe the Future of Life Institute letter points in a wise direction: to take ownership of the pace of technological change.<\/p>\n

Despite what we have seen of the disruption caused by social media, Silicon Valley still tends to follow Facebook\u2019s infamous motto of \u201cmove fast and break things<\/a>\u201d.<\/p>\n

I believe a wise course of action is to slow down and think about where we want to take these technologies, allowing our systems and ourselves to adjust and engage in diverse, thoughtful conversations. It is not about stopping, but rather moving at a sustainable pace of progress. We can choose to steer this technology, rather than assume it has a life of its own that we can\u2019t control.<\/p>\n

After some thought, I have added my name to the list of signatories of the open letter, which the Future of Life Institute says now includes some 50,000 people. Although a six-month moratorium won\u2019t solve everything, it would be useful: it sets the right intention, to prioritise reflection on benefits and risks over uncritical, accelerated, profit-motivated progress.<\/p>\n

<\/p>\n
\n

Watch him never say \u201cwe need more regulation on AI\u201d https:\/\/t.co\/kDNALQkz4G<\/a><\/p>\n

— Avijit Ghosh (@evijitghosh) March 13, 2023<\/a><\/p><\/blockquote>\n