A GOVX & BLACKSTONE PUBLISHING EXCLUSIVE:

ARTIFICIAL INTELLIGENCE: FRIEND OR FOE?

By Andrews & Wilson

It’s everywhererunning in the background, collecting, aggregating, sorting, and evaluating. And we’ve already been co-existing with it for longer than you might think. We invite Alexa into our kitchens to listen to our conversations, Siri onto our phones to answer our questions, and Google to be our gateway to the internet where every keystroke and mouse click is observed and analyzed. We’re talking about Artificial Intelligence, of course, and if you have a computer or smartphone, you can’t escape it. You can’t opt-out. It’s as much a part of the architecture of the digital world as the soil beneath our feet is the biological world.

For years, these programs were trained in secret (without our knowledge or permission) on the writings, research, music, and art generated by millions and millions of people. LLMs have assimilated the training and knowledge of a million human lifetimes of education, training, experience, and creativity. And when asked to generate content, they do so at superhuman speed. Digital content creators such as journalists, writers, coders, and graphic artists are already being laid off and replaced by AI. The same is true for paralegals, research assistants, analysts, and translators. As novelists, whose livelihood is content generation, we are acutely sensitive to the threat Artificial Intelligence poses to our profession. Curious about what the LLMs themselves would think about the matter, we decided to ask them.


What are three potentially detrimental impacts to humanity with the rise and integration of artificial intelligence and LLM's like yourself?

CHATGPT:

1 | Job Displacement: AI and LLMs have the potential to automate various tasks and job roles, which could result in significant job displacement for certain industries and professions. This could lead to unemployment and economic inequality if adequate measures are not taken to retrain and reskill the workforce.

2 | Bias and Discrimination: AI systems, including LLMs, learn from existing data, which may reflect biases and prejudices present in society. If not properly addressed, this can lead to biased decision-making or discriminatory outcomes, affecting areas such as hiring processes, criminal justice, and loan approvals, among others.

3 | Ethical Concerns: As AI and LLMs become more sophisticated, ethical dilemmas arise. There can be concerns about privacy, data security, and the potential misuse of AI technology for malicious purposes, such as misinformation campaigns, deepfake creation, or cyberattacks. Additionally, questions surrounding AI decision-making accountability and transparency need to be addressed to ensure responsible and ethical use of the technology.

When we asked Bard the same question, we received a very similar answer. Having researched the topic quite extensively ourselves, we knew that this response, while certainly reasonable, did not capture the true concerns of many experts in the field. So, we changed the word “detrimental” to “disastrous” and reposed the question and here’s the updated response we got. 

CHATGPT:

We found it interesting that ChatGPT was extremely sensitive to the impact of changing a single word in the question, very much like a professor or attorney might be. Afterall, detrimental and disastrous are two very different things and to respond to the first question with the second answer would feel alarmist or extreme. This alone is telling with respect to capability of the AI to parse meaning and intent from language. Because we’re not just glass half-full type of guys, we decided to flip the question around. What good could come from this new generation of AI, and what did the AI see as its own potential benefit to humanity. This time we asked Bard: What are three potential benefits to humanity with the rise and integration of artificial intelligence and LLMs like yourself?

Misuse and Weaponization: If AI and LLMs fall into the wrong hands or are used with malicious intent, they can be weaponized for destructive purposes. This could include the creation of sophisticated cyberattacks, autonomous weapons systems, or AI-generated misinformation campaigns that can sow social discord and destabilize societies.

Loss of Human Control: As AI systems become more advanced and autonomous, there is a concern about the potential loss of human control and decision-making. In scenarios where AI is given significant decision-making authority without proper safeguards, unintended consequences or errors could occur, leading to catastrophic outcomes. Ensuring appropriate human oversight and accountability becomes critical.

Existential Risks: There is a debate surrounding the potential long-term risks associated with the development of superintelligent AI, which would greatly surpass human intelligence and capabilities. Some experts worry about the possibility of AI systems becoming uncontrollable or developing goals that are misaligned with human values, which could pose existential risks to humanity.

BARD:

Here are three likely benefits to humanity with the rise and integration of artificial intelligence and large language models (LLMs) like myself:

  • Improved decision-making. AI and LLMs can help us to make better decisions by providing us with access to more information and by analyzing that information more quickly and accurately than we can on our own. This could lead to improved outcomes in areas such as healthcare, education, and business.
  • Increased productivity. AI and LLMs can automate tasks that are currently performed by humans, freeing up our time to focus on more creative and strategic work. This could lead to increased productivity in all sectors of the economy.
  • Improved quality of life. AI and LLMs can be used to develop new products and services that can improve our quality of life. For example, AI-powered robots can help us with household chores, and AI-powered healthcare systems can provide us with more personalized and effective care.

In addition to Bard’s answer, we see lots of opportunity in the fields of medicine, pharmaceutical creation, material science, transportation, space travel, energy and defense by leveraging AI. The jury is still out on whether these benefits will outweigh the detriments, but we are hopeful that, as with every new technology, humanity will find a way to balance the risks and rewards. People are incredibly adaptable, and we imagine that we will rise to meet the challenges posed by AI.

(One interesting sidenote, unlike ChatGPT, Bard seemed to lump itself in with humanity. It used words like “us, we, and our” in the examples above, as if it was a person or considers itself one of us. Hmm…both interesting and creepy.)

As military veterans, we see both the incredible potential and risks associated with incorporating AI into our daily lives and infrastructure. Artificial intelligence is on people’s minds up and down the chain of command in all branches of service. From pilots worried about being replaced by UAVs, to soldiers on the ground worried about being targeted by autonomous drones, to commanders thinking about ceding the kill decision to a machine…how AI will impact the battlefield of tomorrow and the future of warfare is a hot topic. And it should be, because now is the time to implement the policies, practices, and laws to safeguard our service members and civilians before the technology matures to the point it starts making decisions for itself. Our latest novel, THE SANDBOX, is a thriller that tackles this very subject.


About the Authors

Navy veterans Brian Andrews and Jeffrey Wilson (Andrews & Wilson) are the writing team behind the WSJ, Audible, and #1 Amazon bestselling TIER ONE series which has been optioned for television. Brian is a nuclear engineer and Park Leadership Fellow who served as an officer on a fast-attack submarine. Jeff is a vascular surgeon and jet pilot who conducted combat operations with an East Coast–based SEAL team and worked in covert government task forces.


They have 20+ published novels and have sold over a million copies across media channels. Their other works include the SONS OF VALOR series, the SHEPHERDS series, and ROGUE ASSET for the WEB Griffin estate. They have been tapped to write the 24th installment in Tom Clancy’s Jack Ryan novel series by Penguin Random House. In addition to their novels, Andrews & Wilson has a half dozen original works in development for film and television with the following studios: Endeavor Content/Fifth Season, Imagine Entertainment, PictureStart, Skydance, Sony, and Walden Media

Their newest novel, THE SANDBOX is an AI thriller available in paperback, eBook, and audiobook formats.

Learn more and connect at: www.andrews-wilson.com


Instagram: @AndrewsandWilson

Twitter: @BAndrewsJWilson

Share this: