{"id":45223,"date":"2024-05-04T18:27:05","date_gmt":"2024-05-04T10:27:05","guid":{"rendered":"https:\/\/wp-productionenv-bjg9h2g2bgg5b8aa.southeastasia-01.azurewebsites.net\/news\/ai-in-space-karpathy-suggests-ai-chatbots-as-interstellar-messengers-to-alien-civilizations\/"},"modified":"2024-05-04T18:27:05","modified_gmt":"2024-05-04T10:27:05","slug":"ai-in-space-karpathy-suggests-ai-chatbots-as-interstellar-messengers-to-alien-civilizations","status":"publish","type":"post","link":"https:\/\/starpath.global\/news\/ai-in-space-karpathy-suggests-ai-chatbots-as-interstellar-messengers-to-alien-civilizations\/","title":{"rendered":"AI in space: Karpathy suggests AI chatbots as interstellar messengers to alien civilizations"},"content":{"rendered":"

On Thursday, renowned AI researcher Andrej Karpathy, formerly of OpenAI and Tesla, tweeted a lighthearted proposal that large language models (LLMs) like the one that runs ChatGPT could one day be modified to operate in or be transmitted to space, potentially to communicate with extraterrestrial life. He said the idea was \u201cjust for fun,\u201d but with his influential profile in the field, the idea may inspire others in the future.<\/p>\n

Karpathy\u2019s bona fides in AI almost speak for themselves, receiving a PhD from Stanford under computer scientist Dr. Fei-Fei Li in 2015. He then became one of the founding members of OpenAI as a research scientist, then served as senior director of AI at Tesla between 2017 and 2022. In 2023, Karpathy rejoined OpenAI for a year, leaving this past February. He\u2019s posted several highly regarded tutorials covering AI concepts on YouTube, and whenever he talks about AI, people listen.<\/p>\n

Most recently, Karpathy has been working on a project called \u201cllm.c\u201d that implements the training process for OpenAI\u2019s 2019 GPT-2 LLM in pure C, dramatically speeding up the process and demonstrating that working with LLMs doesn\u2019t necessarily require complex development environments. The project\u2019s streamlined approach and concise codebase sparked Karpathy\u2019s imagination.<\/p>\n

\u201cMy library llm.c is written in pure C, a very well-known, low-level systems language where you have direct control over the program,\u201d Karpathy told Ars. \u201cThis is in contrast to typical deep learning libraries for training these models, which are written in large, complex code bases. So it is an advantage of llm.c that it is very small and simple, and hence much easier to certify as Space-safe.\u201d<\/p>\n

Our AI ambassador<\/h2>\n

In his playful thought experiment (titled \u201cClearly LLMs must one day run in Space\u201d), Karpathy suggested a two-step plan where, initially, the code for LLMs would be adapted to meet rigorous safety standards, akin to \u201cThe Power of 10 Rules\u201d adopted by NASA for space-bound software.<\/p>\n

This first part he deemed serious: \u201cWe harden llm.c to pass the NASA code standards and style guides, certifying that the code is super safe, safe enough to run in Space,\u201d he wrote in his X post. \u201cLLM training\/inference in principle should be super safe \u2013 it is just one fixed array of floats, and a single, bounded, well-defined loop of dynamics over it. There is no need for memory to grow or shrink in undefined ways, for recursion, or anything like that.\u201d<\/p>\n

,<\/p>\n

That\u2019s important because when software is sent into space, it must operate under strict safety and reliability standards. Karpathy suggests that his code, llm.c, likely meets these requirements because it is designed with simplicity and predictability at its core.<\/p>\n

In step 2, once this LLM was deemed safe for space conditions, it could theoretically be used as our AI ambassador in space, similar to historic initiatives like the Arecibo message (a radio message sent from Earth to the Messier 13 globular cluster in 1974) and Voyager\u2019s Golden Record (two identical gold records sent on the two Voyager spacecraft in 1977). The idea is to package the \u201cweights\u201d of an LLM\u2014essentially the model\u2019s learned parameters\u2014into a binary file that could then \u201cwake up\u201d and interact with any potential alien technology that might decipher it.<\/p>\n

\u201cI envision it as a sci-fi possibility and something interesting to think about,\u201d he told Ars. \u201cThe idea that it is not us that might travel to stars but our AI representatives. Or that the same could be true of other species.\u201d<\/p>\n

\n

\"A<\/p>\n

A technician attaches a gold record to a Voyager space probe, USA, circa 1977. Voyager 1 and its identical sister craft Voyager 2 were launched in 1977 to study the outer Solar System and eventually interstellar space. The record, entitled \u2018The Sounds Of Earth\u2019 contains a selection of recordings of life and culture on Earth. The cover contains instructions for any extraterrestrial being wishing to play the record. <\/p>\n

Credit:
\n Getty Images<\/p>\n<\/figure>\n

Karpathy isn\u2019t the first person to suggest sending LLMs into space. In December 2022, not long after the launch of ChatGPT, University of Vermont robotics professor Josh Bongard tweeted an idea for a sci-fi novel: \u201cWe send ChatGPT into space, with instructions for how ETs may query it to learn about humanity.\u201d<\/p>\n

LLMs represent a highly compressed interactive library of human knowledge that could serve to answer almost any question aliens might have about human civilization (it might also confabulate information, making things up about us where there are gaps in knowledge, but the aliens won\u2019t know the difference).<\/p>\n

In an October 2023 blog post, software developer Lee Mallon imagined sending an LLM into deep space. \u201cThrough this unique exchange, another civilization could get a taste of our culture, our kindness, our artistic expressions, and the leaps and bounds we\u2019ve made in technology and understanding our world,\u201d he wrote. \u201cThey could listen to our poems, read our stories, and maybe even appreciate a good meme or pixel art, seeing the brighter sides of our human nature.\u201d<\/p>\n

,<\/p>\n

Of course, to run the space-bound LLM, the extraterrestrial civilization would need to possess advanced computational capabilities and an understanding of the format in which the model\u2019s weights are encoded. They\u2019d need a compatible hardware architecture capable of processing the matrix operations involved in running the model, as well as a software framework that could interpret the binary file and initialize the LLM. But even if we don\u2019t provide instructions on how to build all of that, a smart enough species might be able to figure it out for themselves.<\/p>\n

\u201cRunning it would, I think, be easy,\u201d Karpathy told us, \u201cbecause the core instructions set could be made very small, e.g., even just a single NAND gate, which is all by itself universal. The bigger problem would be that they will get text as the output, for example in English, and would have to learn themselves how to interpret it or what it means, by interacting with the LLM over a long period of time. It\u2019s not obvious!\u201d<\/p>\n

Despite these challenges, if the extraterrestrial civilization were to decipher and run the LLM successfully, it could open up a new avenue for interstellar communication and knowledge exchange, assuming the model was trained on a range of human knowledge. But given known biases in AI training datasets, we can only imagine the fights that might break out over who gets to decide what goes into the LLM we hypothetically send into space.<\/p>\n

But is it a good idea?<\/h2>\n

The idea of broadcasting or launching an LLM into space raises interesting questions about safety and ethics considerations. Sending messages to alien civilizations, often called active SETI (search for extraterrestrial intelligence), has been a controversial topic for some in the past.<\/p>\n

\n

\"A<\/p>\n

A colorized version of the Arecibo message, a binary transmission beamed into space in 1974.<\/p>\n

Credit:
\n Arne Nordmann \/ Wikipedia<\/p>\n<\/figure>\n

Critics typically argue that deliberately broadcasting our presence and technological capabilities to unknown alien civilizations could potentially attract unwanted attention from hostile entities. That might put our planet or species at risk.<\/p>\n

,<\/p>\n

Karpathy himself is aware of the risks. \u201cOh, absolutely, major safety concerns. See the Three Body Problem<\/em>,\u201d he told us, referring to the famous 2008 novel by Liu Cixin where an extraterrestrial civilization intercepts a message from Earth and subsequently launches an invasion.<\/p>\n

Another risk, aside from a hostile takeover, is that the receiving aliens might also get the wrong idea about us from the LLM if it were trained on a broad spectrum of human culture.<\/p>\n

In his blog post, Lee Mallon wrote, \u201cAs they conversate through the depths of our digital knowledge, they\u2019d also stumble upon the darker chapters of our story. They\u2019d learn about our knack for warfare, our greed, and the many destructive tools we\u2019ve crafted. This could paint a scary picture, making us appear as potential threats. They might start to wonder if giving us a cosmic call is a good idea or a recipe for disaster or if we need to be removed from the chessboard altogether.\u201d<\/p>\n

But even if we trained the LLM solely on humanity\u2019s better attributes, it might not make the best interstellar diplomat due to technical drawbacks and first-contact delicacies.<\/p>\n

\u201cGoodness, the idea of LLMs as our representatives to other species is terrifying,\u201d said Ars Technica Senior Space Editor Eric Berger. \u201cIs it a good idea? God, no. It seems that if you were to chat with an LLM long enough it would start to get hallucinogenic. Or like when Kevin Roose was asked to leave his wife. That sort of thing. Any interaction with a new species would probably be a very delicate thing, requiring an incredible amount of nuance.\u201d<\/p>\n

Putting delicate nuance aside for the sake of the thought experiment, if LLMs are going to serve as our interstellar ambassadors, Karpathy jokes about the importance of sending highly polished code to represent humanity to the broader universe: \u201cMaybe one day we\u2019ll ourselves find LLMs of aliens out there, instead of them directly,\u201d he tweeted. \u201cMaybe the LLMs will find each other. We\u2019d have to make sure the code is really good, otherwise that would be kind of embarrassing.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"

On Thursday, renowned AI researcher Andrej Karpathy, formerly of OpenAI and Tesla, tweeted a lighthearted proposal that large language models (LLMs) like the one that runs ChatGPT could one day be modified to operate in or be transmitted to space, potentially to communicate with extraterrestrial life. He said the idea was \u201cjust for fun,\u201d but […]<\/p>\n","protected":false},"author":1,"featured_media":45224,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":"","_links_to":"","_links_to_target":""},"categories":[2],"tags":[293,9697,5122,4581,9698,9699,9596,9700,9701,9702,9703,9704,9705,5295,4343,21,3086,3110],"class_list":["post-45223","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news","tag-ai","tag-andrej-karpathy","tag-arecibo","tag-chatgpt","tag-chatgtp","tag-fei-fei-li","tag-google-2","tag-google-ai","tag-gpt-2","tag-gpt-3","tag-gpt-4","tag-large-language-models","tag-llms","tag-machine-learning","tag-openai","tag-space","tag-tesla","tag-voyager"],"acf":[],"_links":{"self":[{"href":"https:\/\/starpath.global\/blog\/wp-json\/wp\/v2\/posts\/45223"}],"collection":[{"href":"https:\/\/starpath.global\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/starpath.global\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/starpath.global\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/starpath.global\/blog\/wp-json\/wp\/v2\/comments?post=45223"}],"version-history":[{"count":0,"href":"https:\/\/starpath.global\/blog\/wp-json\/wp\/v2\/posts\/45223\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/starpath.global\/blog\/wp-json\/wp\/v2\/media\/45224"}],"wp:attachment":[{"href":"https:\/\/starpath.global\/blog\/wp-json\/wp\/v2\/media?parent=45223"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/starpath.global\/blog\/wp-json\/wp\/v2\/categories?post=45223"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/starpath.global\/blog\/wp-json\/wp\/v2\/tags?post=45223"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}