Emerging technologies, including AI and quantum computing, hold the potential to revolutionize our world for the better. Yet, as these innovations progress, unsettling reports of AI-assisted bioweapons are becoming increasingly plausible.

AI is already driving positive change, from optimizing energy grids to accelerating the development of clean energy solutions in the battle against climate change. It also plays a crucial role in agriculture, helping farmers enhance crop yields and make informed decisions to boost food production for a growing global population.

However, while many work tirelessly to protect our planet, others appear bent on its destruction. The rise of autonomous weapons systems, AI-guided drones in military operations, and robotic soldiers suggests that our most formidable enemy might be ourselves. But how did we arrive at this perilous crossroads?

From Historical Biowarfare to AI-Assisted Destruction

The history of using viruses and bacteria as weapons is long and grim, with few nations able to claim moral superiority. During World War II, Japan’s notorious Unit 731 deployed typhus and cholera as biological weapons, introducing chemical warfare to the battlefield. Similarly, during the Cold War, the Soviet Union’s bioweapon program produced smallpox, plague, anthrax, and drug-resistant strains of plague.

In the U.S., from 1949 to 1969, the military developed biological weapons using agents like anthrax and botulinum toxin. Although President Richard Nixon terminated the program in 1969, the march of technology is once again leading governments down a treacherous path reminiscent of a darker era.

How AI is Lowering the Bar for Mass Destruction

Recently, biochemist Rocco Casagrande demonstrated to White House officials how AI chatbots, including Gemini, Claude, and ChatGPT, could aid in creating dangerous pathogens. He cautioned that the same AI tools designed to help users find recipes could also be used to craft bioweapons in ordinary basements or even high school labs.

In response, the White House has secured voluntary commitments from leading AI companies to evaluate and manage biosecurity risks associated with AI. This proactive step led to last year’s Global AI Safety Summit in London, where Vice President Kamala Harris highlighted the existential threat posed by AI-enhanced bioweapons.

The Alarming Ease of Creating Bioweapons with AI

A recent study underscores the frightening potential for AI models, originally developed for drug discovery, to be repurposed for bioweapons. Within just six hours, an AI system generated 40,000 potentially toxic molecules, including known chemical warfare agents and new compounds predicted to be even more lethal.

The speed and ease with which such AI tools can be misused raise significant concerns. The study illustrates that creating toxic molecules requires minimal time, effort, and resources, emphasizing the dual-use dilemma of this technology.

Researchers advocate for stricter oversight of AI tools and datasets to prevent their misuse in creating biochemical weapons. Additionally, many call for ethical AI training, particularly for students, to increase awareness of these risks.

The authors also recommend fostering dialogue at scientific conferences, establishing a code of conduct for AI companies, and enhancing security for public-facing APIs to safeguard humanity from its darker impulses. However, the most daunting challenges lie in addressing the more complex issues beyond the scope of a lone individual misusing AI in a basement.

AGI 2029 and the Terminator’s Grim Prediction

Forty years have passed since the movie Terminator warned of a future where AI could surpass human intelligence and take control by 2029. Ironically, in 2024, influential figures like Nvidia CEO Jensen Huang and futurist Ray Kurzweil predict that we’ll achieve artificial general intelligence (AGI) by 2029.

Instead of viewing AI as the villain, perhaps we should question why we’re so startled by our reflection. Biological warfare dates back to 1763-64, when the British used smallpox-contaminated blankets against Indigenous communities.

Is AI the Real Threat, or Is It a Mirror of Our Humanity?

The issue isn’t AI’s technological capabilities but how it amplifies humanity’s worst instincts. Despite my optimism that there are more good people than bad, AI inevitably reflects the data it’s trained on—data created by humans.

If AI appears to promote war, cruelty, intolerance, hatred, greed, and destruction, it merely mirrors our own behaviors. After all, like our AI creations, we are shaped by those who came before us.

Over 100 scientists have committed to policies preventing AI’s use in bioweapons, seeking to alleviate concerns that individuals or malicious groups could develop bioweapons from their basements. Yet history suggests that nation-states are more likely to break the rules and weaponize biological agents.

Instead of taking accountability, some may find it easier to blame AI or any new technology. But it’s humans, not AI, who will create the next pandemic or unleash the biological weapons that threaten our existence.

Breaking this cycle of self-destruction and changing the world for the better is a challenge that lies with humanity, not AI.