Sophia, AGI, and the Ethics of Anthropomorphic Robotics: A Conversation with Ben Goertzel

This podcast clip features Ben Goertzel, chief scientist of Hanson Robotics and co-founder of SingularityNET, discussing the creation and public perception of Sophia the robot. Goertzel contrasts his math-oriented approach to AGI with David Hanson's vision of computational compassion through human-like robots. Both share a goal of creating benevolent AGI that improves life for all sentient beings, with Hanson focusing on emotional connection and Goertzel on abstract cognitive algorithms. The discussion highlights Goertzel's journey to Hong Kong, his involvement with Hanson Robotics, and the eventual co-founding of SingularityNET, a blockchain-based cloud mind platform intended to host Sophia and other robots.

Goertzel delves into the practical realities of developing Sophia, describing her current software as a "walking skeleton" – a simple, integrated system with basic AI components (computer vision, dialog, memory, learning) but no true AGI. He explains the challenges of speech-to-text in diverse accents and the limitations of current neural networks in maintaining coherent narrative structures compared to rule-based systems. A key distinction is made between AI systems that can speak fluently about ungrounded topics (like Sophia, Google Mena, Facebook's chatbot) and a hypothetical AGI-aimed robot toddler whose language would be grounded in its direct experience.

The conversation addresses the controversy surrounding Sophia, particularly the frustration from AI community figures like Yann LeCun regarding the public's tendency to anthropomorphize and over-attribute intelligence to the robot. Goertzel acknowledges this but defends Hanson's artistic approach, arguing that the illusion serves a benevolent purpose: to foster empathy and compassion, which are crucial for biasing future AGIs toward positive outcomes. He points out the irony of critics from companies using AI for potentially destructive ends (like Facebook) while Sophia aims for beneficial user experiences.

Finally, Goertzel emphasizes the ethical considerations of transparency versus the beneficial use of illusion. He states that Hanson Robotics has been transparent about Sophia's underlying systems (timeline editor, chatbot, OpenCog), but people still attribute more agency than is present. He argues that if this anthropomorphism leads to better patient interaction in medical robotics (e.g., elder care, nursing assistants during pandemics), it can be a force for good, provided the technology is used ethically. The core message is that the problem lies not in the technology itself, but in how it is applied, whether for scamming or for compassionate assistance.

"David is the brilliant mind behind this Sophia roba and he remains so far he remains more interesting than his creation although although she may be improving faster than he."

"I wanted to create benevolent AGI that would you know create massively better life for all humans and all sentient beings including animals plants and superhuman beings."

"He wanted to get computational compassion like he wanted to get machines that that would would love people and empathize with people and he thought the way to do that was to make a machine that could know look people either eye face to face look at look at people and make people love the machine and the machine loves the people back."

"It seemed to me that based on our current state of knowledge the best way to bias the AG is we create toward benevolence would be to infuse them with love and compassion the way the way that we do our own children."

"The first principle is the robot is always broken working."

"There's a lot of AI AM there there's not any AGI in there."

"I mean like if you look at the lubna prize and the systems that beat the Turing test currently they're heavily rule-based because like you had said narrative structure to create compelling conversations you currently neural networks cannot do that well."

"Neither one of them really understands what's going on."

"If Sophia just had nothing inside her head said nothing if she just sat there we already prescribed some intelligence to it."

"If your ethics allows you to use machine learning in such a blatantly destructive way why would your ethics not allow you to use machine learning to make a lovable theatrical robot that draws some foolish people into its theatrical illusion?"

"You can tell people exactly how it's working and they won't care they want to be drawn into the illusion and he was 100% 100% correct."

"The thing is even if you get up onstage and tell people how sophia is working and then they talk to her they still contribute more agency and consciousness to her then then is is is really there."