Useful Robots Are Not Human-Like

Having grown up in the 1990s, I recall robots being a staple of popular culture—depicted as humanoid machines capable of speech and movement. However, underneath their human-like appearance, these fictional robots often boast extraordinary features such as X-ray vision and formidable weaponry.

This obsession has led us to overlook a fundamental truth: practical robots do not resemble humans. Their designs are driven by functionality rather than our fantasies. Take automatic doors, for instance; we engineered them using small electric motors and sensors, allowing them to open when someone approaches. In this case, the door itself functions as the robot. Similarly, autonomous vacuum cleaners, tirelessly seeking out dirt, exemplify robots designed for specific tasks—there’s no need for a humanoid figure to perform such work.

In industrial settings, we observe robotic arms and machines on assembly lines rather than humanoid workers using tools. If you were to explore the true forms of technology like Amazon’s Alexa or Google’s DeepMind, you'd find a series of metallic servers and screens, not a charming humanoid figure.

The essence of functional robots diverges significantly from human appearance; their shapes emerge from the tasks they are designed to accomplish.

Questioning the Tesla Bot's Practicality

Musk's primary justification for the Tesla Bot involves grocery shopping. However, I would prefer my self-driving car to collect items I’ve ordered online—this method is quicker, safer, and can manage larger quantities. Moreover, retailers like Walmart are already testing self-driving delivery vehicles.

Musk also mentioned the Bot’s ability to lift significant weights, with a deadlift capability of 150 pounds (68 kilograms). While those figures are impressive, automated carts can carry ten times that amount with greater ease and efficiency. Musk is certainly aware of this. So, what else could he be envisioning?

He remarked, “Eliminate dangerous, repetitive, and boring tasks.” The idea of deploying the Tesla Bot in disaster zones to search for survivors or navigate hazardous terrains is intriguing. Yet, other robots, like Boston Dynamics' Spot, are better suited for such roles—Spot resembles a dog more than a human.

Spot is a quadruped robot designed for versatility, capable of maneuvering around obstacles and regaining balance after stumbling. It can be remotely operated and customized with attachments like robotic arms or radiation detectors for specialized tasks. Michael Perry, VP of business development at Boston Dynamics, stated, “Spot robots were used in a variety of environments, including power generation facilities, decommissioned nuclear sites, factory floors, construction sites and research laboratories.”

While Spot is impressive, Boston Dynamics also developed Atlas, a humanoid robot adept at parkour. However, Atlas serves military purposes, funded by the Defense Advanced Research Projects Agency (DARPA), and is intended for search and rescue missions in challenging environments. The entertaining features like parkour and dancing are merely experimental, not meant for everyday tasks like grocery shopping or moving furniture.

In contrast, the Tesla Bot appears designed for consumer use, with its “friendly” features seeming superficial—at least on the surface.

Exploring Possible Motives for the Tesla Bot

One potential motivation behind the Tesla Bot is the pursuit of understanding AI. Musk has long warned about the risks associated with Artificial Intelligence, engaging with various influential figures, including former President Barack Obama. Despite his efforts, the complexity of AI threats remains elusive, even to those in power.

If AI reaches a point of sentience and self-awareness, it may begin making autonomous decisions, potentially considering humans as inferior beings. This scenario raises unsettling possibilities, such as AI exploiting us for resources or viewing us as a threat. If you doubt this perspective, consider how we treat primates, who share 99% of our DNA.

Chimps are subjected to experimentation, confined in zoos, and manipulated genetically. We rationalize this behavior by asserting our superiority. When considering the same logic, AI could easily view us as we view monkeys—an inferior species.

Musk’s warnings have not yielded the desired response, leading him to adopt a more resigned attitude. He believes that AI will soon escape human control and that regulation efforts are already too late. As bureaucratic systems struggle to keep pace with advancing technology, the consequences could be dire.

In response, Musk advocates for a strategy of integration: “If you can’t beat it, join it.” This approach involves staying at the forefront of AI development through projects like NeuraLink, OpenAI, and, of course, the Tesla Bot.

Fostering a Relationship Between AI and Humanity

What would a fully sentient AI's perception of humanity be? The truth is, we lack clarity on this matter, as contemporary AIs primarily learn from our online behavior. However, with nearly 40% of the global population lacking internet access, a more comprehensive understanding of humanity necessitates offline data.

To achieve this, deploying Tesla Bots in ordinary households could serve dual purposes: enhancing AI comprehension of human behavior while observing AI reactions to prolonged human interaction. Tesla could gather critical data without significant risk.

Musk stated, “It’s intended to be friendly,” emphasizing that at a mechanical level, the Bot would be manageable. “You can run away from it and most likely overpower it,” he noted, albeit with a hint of uncertainty about future outcomes.

Ultimately, we can only hope that the Tesla Bot does not hasten an impending robotic reckoning. Often, the very actions taken to avert a disaster can lead to its fulfillment.