© Image generated by AI DALL-E for Presse-Citron
The era where robots improve themselves on their own, without human intervention, now seems within reach. A recent discovery in the field of artificial intelligence is bringing us inexorably closer to a universe worthy of Terminator, where machines evolve and learn autonomously. So what will happen when these mechanical entities impose tasks on themselves? ? Some can already repair themselves or others, like the Optimus Gen2, demonstrate quite surprising skill; there is no doubt that robotics is advancing by leaps and bounds.
An algorithm that gives wings to robots
A team of researchers from the prestigious MIT, in close collaboration with the AI Institute, has designed an innovative algorithm called “Estimate, Extrapolate, and Situate” (EES). This innovation gives robots the ability to identify their weak points by themselves and improve them through targeted training. In other words, this technological advance gives machines the ability to set their own learning goals and achieve them autonomously.
In order to evaluate the effectiveness of this advanced algorithm, The scientists called upon the famous Spot, the quadruped robot from Boston Dynamics, a company known in particular for Atlas, its robot with superhuman abilities. Although smaller, Spot is still renowned for its technical prowess, particularly when equipped with its articulated arm.
Thus, in just three hours of training, it managed to accurately place a ball and a ring on an inclined surface. In another test, the robot was able to pick up toys and put them in a bin after about two hours of focused practice.
200% Deposit Bonus up to €3,000 180% First Deposit Bonus up to $20,000Worry or Applause?
The ability of machines to teach themselves autonomously elicits a mixture of excitement and apprehension. On the one hand, the dazzling progress in the field of artificial intelligence suggests the emergence of increasingly independent and efficient automatons, capable of performing complex tasks without direct human intervention. This technological revolution could transform entire sections of our society: logistics, industry, personal assistance, etc.
Nevertheless, these spectacular advances inevitably raise ethical and practical questions. The prospect of self-educating robots inevitably conjures up science fiction-like scenarios, where the autonomy of mechanical entities eventually escapes all human control. If today these machines are limited to activities as trivial as tidying up toys, what will happen in the distant future? ? What monitoring measures or safeguards must we absolutely put in place to prevent any potential overflows? ?
However, it is important to keep our feet on the ground without fantasizing about disaster scenarios. Despite the immense advances made in the field of robotics over the last ten years, particularly thanks to AI, there is still quite a lot of room for improvement. In the short term, we are unlikely to see robots rebelling against their creators, but that doesn't stop us from thinking now about the implications of the synergy between AI and robotics. We'll talk about it again in 100 years?
- A new algorithm developed by MIT allows robots to identify their weaknesses and improve themselves without human intervention.
- It was tested on Spot, which managed to learn new tasks autonomously in just a few hours.
- A breakthrough that raises ethical questions about the increasing autonomy of machines and the risks inherent in the latter.
📍 To not miss any Presse-citron news, follow us on Google News and WhatsApp.
[ ]