Is Artificial Intelligence Really a Threat to Humanity?

In short, yes! If you’ve had that eerie feeling looking at AI-powered robots with human faces, you’re not alone. But, have you thought of a humanoid invasion? The current level of AI could definitely lead to mass extinction on the scale of a nuclear war or pandemic, but not like that!

Artificial intelligence has made some groundbreaking ascensions in just a few years and is continually becoming the center of interest in various professional fields.

Taking over jobs, risking mass extinction, leading to social and security issues; Why are we seeing such speculative dilemmas come from all-time leading experts, and what’s alarming Google to have collaborations with the European commission to legislate new AI laws?

To all who say, “Just throw water!”, the situation isn’t as simple as it looks. The dilemmas aren’t speculative, the panic is real, and there’s an impending sense of doom!

This article will discuss 3 major problems with GPT technology and a possibility it could hold for the future of human-kind.

But Before That, Why Should You Care?

After all, we’re not seeing a humanoid invasion or AI revolution. However, the AI-driven attacks aren’t happening on a physical landscape of the world. It’s the political, economical and social attacks that hold the potential to jeopardize the world wide web.

The problem is, Generative Pre-trained Transformer (GPT) is too powerful to be given access to common public. And we’re already seeing chaos through misuse in so many areas.

Here are 3 Problems

With a global uprising in surveillance technology and businesses, especially cameras, there is loads of data everywhere on web. And while this technology can be used to counter terrorism, improve security and give us access to more information, it can also be used in detrimental ways.

#1. GPT-Powered Consumer Manipulation and false propagandas

The use of GPT by authoritative figures in building “vast” and “personalized” marketing infrastructures has already caused cult-like polarization and cultivated extremist views in different parts of the world.

In addition, artificial intelligence can organize and analyse loads of sensitive data to unethically gauge and manipulate consumer action.

Besides exploits in the digital marketing sectors, 100% AI-generated faux social media communities could be put up for social engineering and other malicious purposes. For a real life instance, a researcher once fine-tuned a model on a dataset of 3.3 million 4chan posts, and generated over 300,000 posts filled with hate speech.

OpenAI has proposed a restriction on AI chips in efforts to fight back the mass spread of disinformation leveraging AI models similar to ChatGPT.

#2. AI-Driven Lethal Autonomous Weapon Systems (LAWS)

AI already aids military and defence systems in so many ways; like intelligent analysis, autonomous systems, cybersecurity, training simulations, etc.

The flip side is, autonomous weapons, such as AI-driven armed drones could easily dehumanize weaponry and target individuals selectively or in mass, which could have implications at a nuclear war scale.

#3. Obsolescence of Human Labor

We’ve already covered this in the article on ChatGPT. The Transformer technology has already taken so many jobs in the lower bracket and is moving up the ladder at an alarming rate.

The solution is simple. One can learn to collaborate with the technology to improve efficiency and quality of their craft.

What’s Wrong With AGI

AGI, or artificial general intelligence, is a fictional concept at the moment. So, the narrative is, what if humans build a machine that can improve itself indefinitely? Experts say, if such Narrow AI ever happened in the future, it could be capable to creating political deception at a global scale which’d be too far for humans to decipher.

As we progress rapidly toward building mythical creatures that feed onto human intelligence and aid malicious organizations with its brain power, we need to make immediate legislative amendments to fit all considerations into play. And seeing Google take measures for it now proves that such future may be much nearer than we had expected.

Conclusion

Humans aren’t incompetent enough to not see and improvise along with AI. In contrast, open access to it can transform how an average person interacts with the world, and it’s likely going to get much better with time.

With changing interactions, the world will change, and we’ll find ways to counter issues just as we did before the conception of every new technological advancement took place.