The development of the atomic bomb was driven by the fear that Nazi Germany would develop such a weapon first. Although the science community were aware of the casualty the bombs could cause, they chose to naively believe that the global society can handle the unprecedented weaponry. The immediate concern of winning WWII deprioritized the moral and legal considerations. It was only after the annihilation in Japan that the creators finally understood the fatality it held to the whole human civilization. Yet by then, retreating was out of option. The international security system had to be rebuilt around a technology that threatens the demolition of the entire civilization.
And after about half a decade, a Chinese biologist, He Jiankui, announced he had created the world's first gene-edited human babies. He used CRISPR-Cas9 on manipulating human embryos, violating the Biomedical Regulations that strictly banned editing human genes. His decision was driven by intense competition and a misjudgment that the world would eventually celebrate his "breakthrough". He raced to be the first, hoping to break a new ground-putting the pursuit of progress ahead of the welfare and the ethical contemplation of the children involved. The international community and China eventually condemned him, but the twins are still out there, carrying genetic modifications with effects not yet discovered. A scientist's blind ambition created a biological situation that humanity may have to grapple with for generations.
The pattern seen in the two examples are not unfamiliar. A powerful technology appears, a competitive pressure arises, and developers get locked into a race where stopping to reflect feels like losing. Discussions of ethical and social impacts are pushed back and winning becomes the only question on agenda. By the time the harm becomes visible, the technology is already everywhere and impossible to pull back. The public are left with no other options but to carry all the consequences.
AI development is now following a similar cycle. Even now, companies increasingly frame AI development as a race between countries, leaving no time for anyone to properly examine the aftermath. While showing concerns in public, none are actually slowing the deployments of models. They understand that AI could reshape the human society itself, yet the main argument they keep repeating is that if we pause, the other will win. This is a strikingly similar justification the Manhattan Project and He Jiankui used. The responsibility of big tech, therefore, is not just to build the future, but to stop and ask whether this future is one that the rest of us would actually want to live in.
