What if advanced AI goes into wrong hands? Potential risks and researchers’ concerns explained

With the introduction of ChatGPT, leading tech companies have shown an inclination towards Generative AI and other advanced Artificial Intelligence tools. Advanced AI models are expected to establish a new dimension in the technology sector in the coming years. However, the potential misuse of these advanced AI models can also lead to significant destruction. According to a Reuters report, the Joe Biden government in America is taking steps to protect American AI from China and Russia by creating a new security layer.

The report highlights the concern of government and private sector researchers regarding the use of advanced AI models for aggressive cyber attacks against America or creating powerful biological weapons. Furthermore, these models can also be used for summarizing large amounts of information and generating content, which is a cause of concern for scientists researching advanced AI. The misuse of advanced AI can lead to increased tension not only in America but also across the world.

Deepfakes

One of the most significant concerns with advanced AI is the creation of deepfakes and spreading rumours. Deepfakes, created by advanced AI algorithms, allow cyber criminals to fabricate videos and circulate them on social media. The number of deep fakes has increased significantly in recent years, and many generative AI tools are easily available on the internet, making it easy for anyone to create deepfake content.

OpenAI, Microsoft and Meta AI have introduced tools for image creation through artificial intelligence, which can easily spread rumours through fake photos. Although many countries have policies to stop these fake videos or content on social media, misleading content and videos are still being circulated easily. Sometimes deepfakes look so real that it is challenging for a common man to identify them. Although social media platforms like Facebook, Twitter (X), and YouTube have policies to fight against these rumours, their effectiveness is still questionable.

Biological Weapons

Another concern regarding the misuse of advanced AI is the creation of biological weapons. According to reports, the American Intelligence Community warns that access to advanced AI capabilities could lead to devastation if it falls into the wrong hands. Researchers Gryphone Scientific and Rand Corporation have explained how large amounts of text content related to biology can be generated through computer programs based on Large Language Model (LLM). This information can be easily accessed by terrorists to create biological weapons and make their path easier.

Cyber attacks

Furthermore, advanced AI models can provide information equivalent to that of a doctor, and viruses that spread infection can be created. Cyber criminals can also use advanced AI to create new tools that can carry out large-scale cyber attacks, enough to destroy large and complex infrastructure such as railways and pipelines. DHS says that China and its allies are working on advanced AI technology that can penetrate America’s cyber defense and carry out major attacks.

In conclusion, advanced AI models have the potential to revolutionize the technology sector, but their misuse can lead to significant destruction. It is crucial to take steps to protect American AI from foreign threats and ensure that these models are used ethically and responsibly.

ALSO READ: What is Digital House Arrest and how to avoid it? | Explained

Inputs from Reuters