A Google software engineer was\u00a0suspended\u00a0after going public with his claims of encountering \u201csentient\u201d artificial intelligence on the company\u2019s servers \u2014\u00a0spurring a debate about how and whether AI can achieve consciousness. Researchers say it\u2019s an unfortunate distraction from more pressing issues in the industry.<\/p>\n
The engineer, Blake Lemoine, said he believed that Google’s AI chatbot was capable of expressing human emotion, raising ethical issues.<\/p>\n
Google put him on leave<\/a><\/strong> for sharing confidential information and said his concerns had no basis in fact \u2014 a view widely held in the AI community.<\/p>\n What\u2019s more important, researchers say, is addressing issues like whether AI can engender real-world harm and prejudice, whether actual humans are exploited in the training of AI, and how the major technology companies act as gatekeepers of the development of the tech.<\/p>\n Lemoine\u2019s stance may also make it easier for tech companies to abdicate responsibility for AI-driven decisions, said Emily Bender, a professor of computational linguistics at the University of Washington.<\/p>\n \u201cLots of effort has been put into this sideshow,\u201d she said. \u201cThe problem is, the more this technology gets sold as artificial intelligence \u2014 let alone something sentient \u2014 the more people are willing to go along with AI systems\u201d that can cause real-world harm.<\/p>\n Bender pointed to examples in job hiring and grading students, which can carry embedded prejudice depending on what data sets were used to train the AI.<\/p>\n If the focus is on the system\u2019s apparent sentience, Bender said, it creates a distance from the AI creators\u2019 direct responsibility for any flaws or biases in the programs.<\/p>\n The Washington Post on Saturday ran an\u00a0interview<\/a><\/strong> with Lemoine, who conversed with an AI system called LaMDA, or Language Models for Dialogue Applications, a framework that Google uses to build specialized chatbots.<\/p>\n The system has been trained on trillions of words from the internet in order to mimic human conversation.<\/p>\n In his conversation with the chatbot, Lemoine said he concluded that the AI was a sentient being that should have its own rights.<\/p>\n He said the feeling was not scientific, but religious: \u201cwho am I to tell God where he can and can\u2019t put souls?\u201d he said on Twitter<\/a><\/strong>.<\/p>\n Alphabet Inc.\u2019s Google employees were largely silent in internal channels besides Memegen, where Google employees shared a few bland memes, according to a person familiar with the matter.<\/p>\n But throughout the weekend and on Monday, researchers pushed back on the notion that the AI was truly sentient, saying the evidence only indicated a highly capable system of human mimicry, not sentience itself.<\/p>\n \u201cIt is mimicking perceptions or feelings from the training data it was given \u2014 smartly and specifically designed to seem like it understands,\u201d said Jana Eggers, the chief executive officer of the AI startup Nara Logics.<\/p>\n The architecture of LaMDA \u201csimply doesn\u2019t support some key capabilities of human-like consciousness,\u201d said Max Kreminski, a researcher at the University of California, Santa Cruz, who studies computational media.<\/p>\n If LaMDA is like other large language models, he said, it wouldn\u2019t learn from its interactions with human users because \u201cthe neural network weights of the deployed model are frozen.\u201d<\/p>\n It would also have no other form of long-term storage that it could write information to, meaning it wouldn\u2019t be able to \u201cthink\u201d in the background.<\/p>\n In a response to Lemoine\u2019s claims, Google said that LaMDA can follow along with prompts and leading questions, giving it an appearance of being able to riff on any topic.<\/p>\n \u201cOur team \u2014 including ethicists and technologists \u2014 has reviewed Blake’s concerns per our AI Principles and have informed him that the evidence does not support his claims,\u201d said Chris Pappas, a Google spokesperson.<\/p>\n \u201cHundreds of researchers and engineers have conversed with LaMDA and we are not aware of anyone else making the wide-ranging assertions, or anthropomorphizing LaMDA, the way Blake has.\u201d<\/p>\n The debate over sentience in robots has been carried out alongside science fiction portrayal in popular culture, in stories and movies with AI romantic partners or AI villains. So the debate had an easy path to the mainstream.<\/p>\n \u201cInstead of discussing the harms of these companies,\u201d such as sexism, racism and centralization of power created by these AI\u00a0 systems, everyone \u201cspent the whole weekend discussing sentience,\u201d Timnit Gebru, formerly co-lead of Google\u2019s ethical AI group,\u00a0said on Twitter<\/a><\/strong>.<\/p>\n \u201cDerailing mission accomplished.\u201d<\/p>\n The earliest chatbots of the 1960s and \u201970s, including ELIZA and PARRY, generated headlines for their ability to be conversational with humans.<\/p>\n In more recent years, the GPT-3 language model from OpenAI, the lab founded by Tesla CEO Elon Musk and others, has demonstrated even more cutting-edge abilities, including the ability to read and write.<\/p>\n But from a scientific perspective, there is no evidence that human intelligence or consciousness are embedded in these systems, said Bart Selman, a professor of computer science at Cornell University who studies artificial intelligence. LaMDA, he said, \u201cis just another example in this long history.\u201d<\/p>\n In fact, AI systems don\u2019t currently reason about the effects of their answers or behaviors on people or society, said Mark Riedl, a professor and researcher at the Georgia Institute of Technology. And that\u2019s a vulnerability of the technology.<\/p>\n \u201cAn AI system may not be toxic or have prejudicial bias but still not understand it may be inappropriate to talk about suicide or violence in some circumstances,\u201d Riedl said.<\/p>\n \u201cThe research is still immature and ongoing, even as there is a rush to deployment.\u201d<\/p>\n Technology companies like Google and Meta Platforms Inc. also deploy AI to moderate content on their enormous platforms \u2014 yet plenty of toxic language and posts can still slip through their automated systems.<\/p>\n In order to mitigate the shortcomings of those systems, the companies must employ hundreds of thousands of human moderators in order to ensure that hate speech, misinformation and extremist content on these platforms are properly labeled and moderated, and even then the companies are often deficient.<\/p>\n The focus on AI sentience \u201cfurther hides\u201d the existence and in some cases, the reportedly\u00a0inhumane working conditions<\/a><\/strong>\u00a0of these laborers, said the University of Washington\u2019s Bender.<\/p>\n It also obfuscates the chain of responsibility when AI systems make mistakes.<\/p>\n In a now-famous blunder of its AI technology, Google in 2015 issued a public apology<\/a><\/strong> after the company\u2019s Photos service was found to be mistakenly labeling photos of a Black software developer and his friend as \u201cgorillas.\u201d<\/p>\n As many as three years later, the company admitted its fix was not an improvement to the underlying AI system; instead it\u00a0erased<\/a><\/strong>\u00a0all results for the search terms \u201cgorilla,\u201d \u201cchimp,\u201d and \u201cmonkey.\u201d<\/p>\n Putting an emphasis on AI sentience would have given Google the leeway to blame the issue on the intelligent AI making such a decision, Bender said.<\/p>\n \u201cThe company could say, \u2018Oh, the software made a mistake,\u2019\u201d she said. \u201cWell no, your company created that software. You are accountable for that mistake. And the discourse about sentience muddies that in bad ways.\u201d<\/p>\n AI not only provides a way for humans to abdicate their responsibility for making fair decisions to a machine, it often simply replicates the systemic biases of the data on which it is trained, said Laura Edelson, a computer scientist at New York University.<\/p>\n In 2016, ProPublica published a\u00a0sweeping investigation into COMPAS<\/a><\/strong>, an algorithm used by judges, probation and parole officers to assess a criminal defendant\u2019s likelihood to re-offend.<\/p>\n The investigation found that the algorithm systemically predicted that Black people were at \u201chigher risk\u201d of committing other crimes, even if their records bore out that they did not actually do so.<\/p>\n \u201cSystems like that tech-wash our systemic biases,\u201d said Edelson. \u201cThey replicate those biases but put them into the black box of \u2018the algorithm\u2019 which can\u2019t be questioned or challenged.\u201d<\/p>\n And, researchers said, because Google\u2019s LaMDA technology is not open to outside researchers, the public and other computer scientists can only respond to what they are told by Google or through the information released by Lemoine.<\/p>\n \u201cIt needs to be accessible by researchers outside of Google in order to advance more research in more diverse ways,\u201d Riedl said.<\/p>\n \u201cThe more voices, the more diversity of research questions, the more possibility of new breakthroughs. This is in addition to the importance of diversity of racial, sexual, and lived experiences, which are currently lacking in many large tech companies.\u201d<\/p>\n Artificial intelligence researchers say the debate over whether Google’s LaMDA chatbot is sentient is an unfortunate distraction from more pressing issues in the industry.<\/p>\n","protected":false},"author":341034,"featured_media":430570,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[16],"tags":[35793,79012,79084,167,79080,79078,79082],"class_list":["post-448954","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-software","tag-artificial-intelligence-ai","tag-blake-lemoine","tag-chris-pappas","tag-google","tag-jana-eggers","tag-language-models-for-dialogue-applications-lamda","tag-nara-logics"],"_links":{"self":[{"href":"https:\/\/mybroadband.co.za\/news\/wp-json\/wp\/v2\/posts\/448954"}],"collection":[{"href":"https:\/\/mybroadband.co.za\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mybroadband.co.za\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mybroadband.co.za\/news\/wp-json\/wp\/v2\/users\/341034"}],"replies":[{"embeddable":true,"href":"https:\/\/mybroadband.co.za\/news\/wp-json\/wp\/v2\/comments?post=448954"}],"version-history":[{"count":0,"href":"https:\/\/mybroadband.co.za\/news\/wp-json\/wp\/v2\/posts\/448954\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/mybroadband.co.za\/news\/wp-json\/wp\/v2\/media\/430570"}],"wp:attachment":[{"href":"https:\/\/mybroadband.co.za\/news\/wp-json\/wp\/v2\/media?parent=448954"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mybroadband.co.za\/news\/wp-json\/wp\/v2\/categories?post=448954"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mybroadband.co.za\/news\/wp-json\/wp\/v2\/tags?post=448954"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
\nNow read:\u00a0Google suspends engineer who claimed company created “sentient” AI<\/a><\/h3>\n","protected":false},"excerpt":{"rendered":"