Cybersecurity starts with people in the digital trenches. Photo courtesy of FBI.
Editor's Note: The is the third in a three-part series.
On June 13, 1863, English writer Samuel Butler sent a lengthy letter to the editor of The Press newspaper in Christchurch, New Zealand. It took up nearly two full columns and addressed a subject that didn’t seem to belong in a 19th-century world: that humans were in danger of being replaced by machines.
In part of his letter, Butler wrote: ‘Day by day…the machines are gaining ground upon us: day by day, we are becoming more subservient to them. The time will come when the machines will hold the real supremacy over the world and its inhabitants.’
https://darwin-online.org.uk/converted/pdf/1935_Butler_Press_CUL-DAR226.1.132a.pdf
While that time wouldn’t start rumbling until the mid-1950s, Butler’s letter is now considered the first documented idea of Artificial Intelligence – now a principal aid in both global computer security and hacking.
And while the vision he describes in 1863 hasn’t yet reached that point, cyber professionals are warning in the same fashion that in order to stop attacks and breaches, humans in the near future will either have to create fool-proof protection, or design computer systems to do it – which would mean building mechanical devices to replace humans in the ongoing cybersecurity war.
“The recent Spartanburg County, S.C., cyber attack should be viewed as more than another local government breach,” said Dr. Noah Schiffman, co-owner of Secursion.ai, which helps companies strengthen their cyber defenses. “It is a warning about the widening gap between the speed of technological change and our ability to secure it.”
The June attack of county offices shut down its IT systems for 17 days. Since then, there has been no word from county officials, spokesmen, or law enforcement officials as to how the breach happened, leads on potential suspects, what was compromised, how many people could potentially be affected, and whether anything was stolen.
It was the county’s third breach in five years – one that Schiffman indicated is likely to repeat if stronger measures aren’t employed because this breach “reflects a much larger problem.”
“Cybersecurity is inherently asymmetric,” he said. “Defenders must continuously protect thousands of users, systems, identities, endpoints, and data flows. An attacker only needs one exploitable weakness. And AI is changing that equation.”
Changes in Governance
One change that will likely have to happen with Spartanburg County, as well as a host of other state government bodies and industry, is how defenses are set up and monitored.
For example, Schiffman pointed out that in 2015, many security programs were organized around perimeter defense: firewalls, antivirus software, intrusion detection, and periodic compliance assessments. The underlying assumption was that organizations could establish a trusted internal network and concentrate their defenses at its boundaries.
But this model no longer reflects operational reality because cloud computing, remote work, mobile devices, SaaS platforms, APIs, Internet-connected operational technology, and globally distributed supply chains “have dissolved the traditional perimeter.”
“Security operations became far faster at finding intrusions, yet attackers became faster at monetizing them,” he said. “The decade therefore shifted the control objective from keeping adversaries outside a trusted perimeter to limiting identity abuse, exposure windows, blast radius, supplier risk, and recovery time.”
Public unrest over personal security has also changed, for the worse: In 2025 alone, the number of cyber crimes reported to the FBI totaled just over 1 million – more than triple the number in 2015. And the number of U.S. complaints have increased from a few thousand per month to almost 3,000 per day.
https://www.ic3.gov/AnnualReport/Reports/2025_IC3Report.pdf
All groups in South Carolina reported $264 million in losses, compared to $146 million in 2024, an 80% increase in only a single year.
But different measures are underway. Schiffman noted that state government has established new guidelines and directions that include two key items:
• An AI Center of Excellence and a statewide AI strategy designed to help agencies evaluate, govern, and responsibly deploy AI technologies.
• Workforce development with programs such as the Palmetto AI Corridor, the University of South Carolina's AI initiatives, and new AI-focused Career and Technical Education pathways, all of which are helping to prepare students and workers for an economy in which AI literacy may become as fundamental as digital literacy.
“Over the past year or two, South Carolina has quietly emerged as one of the more proactive states in developing an AI strategy,” Schiffman said. “States that succeed in the AI era…will develop stronger talent pipelines, foster innovation, build public trust, and create environments where businesses can invest and grow with confidence. South Carolina appears to be positioning itself to do exactly that with AI.”
The role of AI
On Sept. 2, 1955, in a letter to the Rockefeller Foundation, Dartmouth College mathematics professor John McCarthy requested project funding for he and nine colleagues to study the idea that a machine can be made to simulate intelligence and learning. He titled his letter ‘Dartmouth Summer Research on Artificial Intelligence’ – the first time the term was used.
While the study did happen over that next summer as intended, funding was only half of what McCarthy had requested and the project ended with a sort of whimper and nothing further.
https://raysolomonoff.com/dartmouth/boxa/dart564props.pdf
AND
https://raysolomonoff.com/dartmouth/boxa/dart564props.pdf
But now, because of McCarthy and this otherwise long-forgotten conference, AI is not only part of daily life, it has already developed into models to generate text, images, videos, audio, software code or other forms of data – known as Generative AI.
It has also created the pejorative term ‘Artificial Idiocy’ in that systems are not fully informed, create fake videos and claims, and doesn’t contain the guts and depth that come from legitimate human interaction.
What it does have, however, is the ability to think mechanically – and, in the world of cybersecurity, aid humans in stopping hackers.
To this end, officials with USC Spartanburg announced this August a partnership with Tranquility AI, a leading provider of AI-powered investigative technology, to launch the nation’s first initiative integrating AI-powered evidence analysis technology across criminal justice education, a regional crime analysis center, and student-led cold case investigations.
The intention is to provide local law enforcement agencies with advanced investigative capabilities while giving students hands-on experience using the same technology used by public safety agencies in real criminal investigations.
“Demand for this work is ever increasing with changes in the technology and the introduction of generative AI in 2022,” said Johnny Williams, senior information security engineer with the division of information technology at USC Upstate.
Brad Bowers, director of law enforcement, strategic solutions at Tranquility AI, added, “Law enforcement agencies are managing more digital evidence than ever before. And the way investigators and analysts are trained must evolve alongside it."
Permanent Solutions?
Of course, the lingering question remains: Will all of the cyber professionals, training, and employment of AI eventually beat the hackers? And if so, will they be enough to simultaneously keep humans in control of the machines that Butler warned of during the reign of England’s Queen Victoria?
Schiffman likened it to trying to predict the future: tricky at best, with the outcome still heavily in doubt.
“Questions about ethics, workforce transformation, privacy, cybersecurity, intellectual property, misinformation, and critical thinking are increasingly becoming part of the discussion,” he clarified. “Those may ultimately prove to be just as important as the technology itself.”
Butler’s solution, printed in his letter in 1863, was to instantly engage the machines in a “war to the death.”
“Every machine of every sort should be destroyed by the well-wisher of his species. Let there be no exceptions made, no quarter shown: let us go back at once to the primeval conditions of the race. If it be urges that this is impossible under the present condition of human affairs, this at once proves…that our servitude has been commenced.”
His solution, of course, is not only impossible, but even more unviable now than in his lifetime.
But Schiffman said it might not necessarily come to that extreme: that from the largest Wall Street firm, industry of all sizes, government security offices, and hospitals systems right down to the door of each personal home computer user, the most effective part of a defense remains simple and straightforward: slow down, verify each sender of information, and never authorize a payment based solely on an email.
“Organizations need strong, layered security measures – and should fully, carefully, validate all users, documents, requests, messages, instructions and payment-verification procedures,” he clarified. “Individuals should independently confirm financial instructions through an official website or verified telephone number. Proper operational security policies need to be drafted and followed at all times…. Cyber crime happens a whole lot more than is publicized. It’s a cat-and-mouse game now, and no one knows where it’s going.”
