Views: 1
Artificial intelligence is advancing at a speed that sometimes outpaces our ability to react. Recently, the technology sector has been shaken by a name that resonates strongly in Silicon Valley offices: Mythos. This new model from Anthropic has set off alarm bells, not because of a lack of power, but quite the opposite.
After completing its development, Anthropic engineers discovered that Mythos possesses an unprecedented capacity to generate hyper-realistic disinformation, breach cybersecurity systems, and draft harmful instructions with surgical precision. It is, in essence, such an effective tool that, in the wrong hands, it becomes a global threat.
Faced with this scenario, the company has made an unusual decision: not to release the model to the public. Currently, Mythos is under lock and key, being analyzed by a coalition of tech giants (Google, Microsoft, Apple, Nvidia, and others) in what has been dubbed Project Glasswing. The goal is to understand how to contain a technology that seems to have crossed the line of what is manageable, and I suppose, how to control it.
What we are experiencing with Mythos is the confirmation of experts' worst fears. We are not facing a common software bug; we are facing the first real warning of the dangers of unregulated AI. It is a real first warning.
While users enjoy the benefits of AI for creative or work tasks, behind the scenes, cybercriminal groups and nations with ulterior motives are already using these models for social engineering attacks and mass surveillance. The risk is not that AI will "awaken," but that it will be used as the ultimate weapon by those who do not seek the common good.
Herein lies the real problem: Is anyone in charge? Currently, humanity's security seems to depend on the ethics of a handful of private companies with very specific interests. However, the control of a technology that can disrupt democracies and collapse infrastructure should not rest solely with technologists, but with society as a whole, through its legal representatives and governments.
If we fail to react after this scare, we will be complicit in the consequences. Governments are distracted by trade wars and other conflicts, while the future is being decided in closed laboratories.
AI is an extraordinary tool, but without a firm, global regulatory framework, the result could be catastrophic. Mythos is a warning. Let's hope it's not the last one we can afford to ignore. I would love to be wrong, but time is running out, and technology waits for no one.
EN 
ES