Advances in artificial intelligence (AI) models, such as those recently announced by Anthropic, are changing the global cybersecurity landscape. This is because it allows not only professional hackers, but also people without advanced programming knowledge, to carry out sophisticated attacks against the entire computing ecosystem that dominates the world.
This is echoed by Professor Robinson Rivas of the Department of Computing at the Central University of Venezuela (UCV), who says that platforms such as Fable 5, the latest innovation from AI company Anthropic, These will facilitate the detection of gaps in global technological infrastructure.
“The new tools (…) what they’re trying to do is create more gaps and vulnerabilities across the global computing infrastructure, which of course includes cryptocurrencies,” Rivas says.
In this scenario, Bitcoin (BTC) and cryptocurrency exchanges emerge as prime targets for cybercriminals due to the sensitive data they store.
In a conversation with CriptoNoticias, Rivas explains that, unlike the underlying Bitcoin protocol, its blockchain has demonstrated mathematical and technical robustness that can resist direct attacks. The weak link in the ecosystem lies with financial services companies. Manage your funds centrally.
This scholar explains: “Most of the user’s cryptocurrency infrastructure is in business-type wallets like Binance or Poloniex, or their large sites. And those sites have centralized infrastructure that connects to the blockchain. So it’s definitely the centralized infrastructure sites that are the focus of attacks using these kinds of tools.
For Rivas, the focus of computer attackers will be on compromising the exchange’s servers rather than modifying the distributed registry. Experts from UCV elaborate: “Many of these sites store their users’ private keys in their infrastructure. And if they can steal or access the users’ private keys in their infrastructure, they have no problem accessing the blockchain.”
This attack pattern is reminiscent of historic incidents in the industry, with the professor recalling attacks on Japan’s Mt. Gox and other large platforms such as Bybit in 2025 and Binance’s BNB Chain network in 2022.
“They (hackers) did not attack the blockchain directly, but crypto-based banking sites. And we will definitely see an increase in these types of attacks,” warns the decentralized systems guru.
Rivas’ comments follow. Anthropic releases cutting-edge AI to market. This is the Claude Fable 5, announced on June 9th, and is the first model of the Mythos class available to the public. As reported by CriptoNoticias, Anthropic noted that the capacity of this model exceeds all models it has released to date.
Filter ineffectiveness and conversational deception
The current debate over the security of tools like Fable 5 and the restrictive government use model known as Mythos primarily revolves around the ethical barriers introduced by its creators.
Although companies like Anthropic limit cybersecurity consultations, Rivas asserts that “there is already evidence that they have always found some way to trick the artificial intelligence filters and gain access to things that are prohibited. It’s happened over and over again.”
According to Rivas, attackers can evade blocks by disguising their intent within the flow of conversations with interactive digital assets. Rivas comments, “With a series of questions, it is very likely that you will gain trust and respond with the belief that your system is not in breach of cybersecurity.”
Similarly, it explains that the use of social engineering targeting legitimate operators represents a direct access route: Neutralizes all technical filters. He emphasizes that if access to the model is limited to state agencies, “another way to access that information is to hack or social engineer the officials who are using the tools.”
According to experts, this systemic vulnerability shows that “none of these platforms are perfect and only persistent monitoring can help here.”
The urgency of state surveillance and a global regulatory framework
The proliferation of these tools not only puts crypto funds at risk, but also privacy and digital identity management on a global level. “The sad reality is that we have already reached the point where national intelligence agencies of all countries, large and small, are conducting surveillance on their citizens,” Rivas said.
“There is far too much evidence of that, without the slightest doubt,” he stresses. This highlights the importance of civil society promoting strong data protection laws. Similar to the European Union and the United Kingdom.
Indeed, this concern about mass surveillance and the leakage of sensitive data is based on a precedent that has had significant international impact. An iconic case is the PRISM program, which came to light in 2013 and demonstrated how intelligence agencies had direct access to information. To servers of major technology companies Collect communications from the public on a global scale.
Similarly, the Cambridge Analytica scandal revealed the use of social engineering techniques and the fraudulent collection of psychological profiles. manipulate the political processThis illustrates the vulnerability of digital identities to automated analysis tools.
Now, instead of proposing a halt to the development of artificial intelligence, he is advocating regulating the liability of technology companies, rather than measures he says are wasteful. “There’s no point in stopping the development of these things,” the expert says.
Rivas argues that There must be a “very solid and clear” regulatory framework Regarding the civil and criminal liability of AI companies, he argued that “AI companies are no longer innocent tools.”
“If third parties use these tools, and these tools are not carefully designed, these companies can also be held liable,” he warns.
Robinson Rivas says the ultimate goal must be a controlled growth of technology that draws historical parallels with nuclear power and respects human rights (HR).
“The great powers did not stop their nuclear development, and no country stopped their nuclear development, but they reached an agreement and, through pressure from civil society and countries, managed to prevent these materials from being used in a more or less responsible way, to limit their use, to somehow limit the growth of technology within certain conditions.” I think that is the moment that will happen with artificial intelligence. “Again, not to stop or censor, but to allow for a more controlled form of growth,” he concluded.
(Tag Translation) Featured
