Harnessing the Power of AI: Why Leashes are Better than Guardrails for Regulation

For years, policymakers have debated the best way to regulate Artificial Intelligence (AI) to prevent its potential risks. A new paper by experts Cary Coglianese and Colton R. Crum proposes a game-changing approach: rather than imposing strict “guardrails” to control AI development, they suggest using flexible “leashes.” This management-based regulation would allow firms to innovate while ensuring public safety.

The authors argue that guardrails are not effective for AI due to its rapidly evolving nature and diverse applications. Social media, chatbots, autonomous vehicles, precision medicine, and fintech investment advisors are just a few examples of how AI is transforming industries. While offering numerous benefits, such as improved cancer detection, AI also poses risks like AV collisions, social media-induced suicides, and bias in digital formats.

Coglianese and Crum provide three case studies illustrating the potential harm from unregulated AI:

1. Autonomous vehicle (AV) crashes
2. Social media-related suicides
3. Bias and discrimination through AI-generated content

In each scenario, firms using AI tools would be expected to put their technology on a leash by implementing internal systems to mitigate potential harms. This flexible approach allows for technological innovation while ensuring that companies are accountable for the consequences of their actions.

Management-based regulation offers several advantages over guardrails:

* It can flexibly respond to AI’s novel uses and problems
* It enables technological exploration, discovery, and change
* It provides a tethered structure that helps prevent AI from “running away”

By embracing this leash-like approach, policymakers can harness the power of AI while minimizing its risks.

The University of Nebraska-Lincoln engineering team has made significant strides in developing soft robotics and wearable systems inspired by human and plant skin’s ability to self-heal injuries. Led by engineer Eric Markvicka, the team presented a groundbreaking paper at the IEEE International Conference on Robotics and Automation that showcased their innovative approach to creating an intelligent, self-healing artificial muscle.

The team’s strategy overcomes the long-standing problem of replicating traditional rigid systems using soft materials and incorporating nature-inspired design principles. Their multi-layer architecture enables the system to identify damage, pinpoint its location, and autonomously initiate a self-repair mechanism – all without external intervention.

The “muscle” or actuator features three layers: a damage detection layer composed of liquid metal microdroplets embedded in silicone elastomer, a self-healing component that uses thermoplastic elastomer to seal the wound, and an actuation layer that kick-starts the muscle’s motion when pressurized with water.

To begin the process, the team induces monitoring currents across the damage detection layer, which triggers formation of an electrical network between traces. Puncture or pressure damage causes this network to form, allowing the system to recognize and respond to the damage.

The next step is using electromigration – a phenomenon traditionally viewed as a hindrance in metallic circuits – to erase the newly formed electrical footprint. By further ramping up the current, the team can induce electromigration and thermal failure mechanisms that reset the damage detection network, effectively completing one cycle of damage and repair.

This breakthrough has far-reaching implications for various industries, particularly in agricultural states where robotics systems frequently encounter sharp objects. It could also revolutionize wearable health monitoring devices that must withstand daily wear and tear.

The technology has the potential to transform society more broadly by reducing electronic waste and mitigating environmental harm caused by consumer-based electronics’ short lifespans. Most consumer electronics have a lifespan of only one or two years, contributing billions of pounds of toxic waste each year.

“If we can begin to create materials that are able to passably and autonomously detect when damage has happened, and then initiate these self-repair mechanisms, it would really be transformative,” Markvicka said.

In the intricate language of mathematics, there lies a fascinating phenomenon known as gauge freedoms. This seemingly abstract concept may seem far removed from our everyday lives, but its impact is felt deeply in the realm of biological sciences. Researchers at Cold Spring Harbor Laboratory (CSHL) have made groundbreaking strides in understanding and harnessing this power.

Gauge freedoms are essentially the mathematical equivalent of having multiple ways to describe a single truth. In science, when modeling complex systems like DNA or protein sequences, different parameters can result in identical predictions. This phenomenon is crucial in fields like electromagnetism and quantum mechanics. However, until now, computational biologists have had to employ various ad hoc methods to account for gauge freedoms, rather than tackling them directly.

CSHL’s Associate Professor Justin Kinney, along with colleague David McCandlish, led a team that aimed to change this. They developed a unified theory for handling gauge freedoms in biological models. This breakthrough could revolutionize applications across multiple fields, from plant breeding to drug development.

Gauge freedoms are ubiquitous in computational biology, says Prof. Kinney. “Historically, they’ve been dealt with as annoying technicalities.” However, through their research, the team has shown that understanding and systematically addressing these freedoms can lead to more accurate and faster analysis of complex genetic datasets.

Their new mathematical theory provides efficient formulas for a wide range of biological applications. These formulas will empower scientists to interpret research results with greater confidence and speed. Furthermore, the researchers have published a companion paper revealing where gauge freedoms originate – in symmetries present within real biological sequences.

As Prof. McCandlish notes, “We prove that gauge freedoms are necessary to interpret the contributions of particular genetic sequences.” This finding underscores the significance of understanding gauge freedoms not just as a theoretical concept but also as a fundamental requirement for advancing future research in agriculture, drug discovery, and beyond.

This rewritten article aims to clarify complex scientific concepts for a broader audience while maintaining the original message’s integrity.