Artificial Intelligence (AI): The risk of over-reliance on quantifiable data

The rise of interest in artificial intelligence and machine learning has a flip side. It might not be so smart if we fail to design the methods correctly. A question out there — can we compress the reality into measurable numbers? Artificial Intelligence relies on what can be measured and quantified, risking an over-reliance on measurable knowledge.

The problem with many other technical problems is that it all ends with humans that design and assess according to their own perceived reality. The designers’ bias, perceived reality, weltanschauung, and outlook — everything goes into the design. The limitations are not on the machine side; the humans are far more limiting. Even if the machines learn from a point forward, it is still a human that stake out the starting point and the initial landscape.

Quantifiable data has historically served America well; it was a part of the American boom after World War II when America was one of the first countries that took a scientific look on how to improve, streamline and increase production utilizing fewer resources and manpower.

Numbers have also misled. Vietnam-era Secretary of Defense Robert McNamara used the numbers to tell how to win the Vietnam War, which clearly indicated how to reach a decisive military victory — according to the numbers.

In a post-Vietnam book titled “The War Managers,” retired Army general Donald Kinnard visualized the almost bizarre world of seeking to fight the war through quantification and statistics. Kinnard, who later taught at the National Defense University, surveyed fellow generals that had served in Vietnam about the actual support for these methods. These generals considered the concept of assessing the progress in the war by body counts as useless; only two percent of the surveyed generals saw any value in this practice.

Why were the Americans counting bodies? It is likely because it was quantifiable and measurable. It is a common error in research design to seek the variables that produce easily accessible quantifiable results, and McNamara was at that time almost obsessed with numbers and the predictive power of numbers. McNamara was not the only one.

In 1939, the Nazi-German foreign minister Ribbentrop, together with the German High Command, studied and measured the French and British war preparations and ability to mobilize. The Germans quantified assessment was that the Allies were unable to engage in a full-scale war on short notice and the Germans believed that the numbers were identical with the factual reality — the Allies would not go to war over Poland because they were not ready nor able. So Germany invaded Poland on the 1st of September 1939 and started WWII.

The quantifiable assessment was correct and lead to Dunkirk, but the grander assessment was off and underestimated the British and French will to take on the fight, which led to at least 50 million dead, half of Europe behind the Soviet Iron Curtain and the destruction of their own regime. Britain’s willingness to fight to the end, their ability to convince the U.S. to provide resources, and the subsequent events were never captured in the data. The German quantified assessment was a snapshot of the British and French war preparations in the summer of 1939 — nothing else.

Artificial intelligence depends upon the numbers we feed it. The potential failure is hidden in selecting, assessing, designing and extracting the numbers to feed artificial intelligence. The risk for grave errors in decision-making, escalation, and avoidable human suffering and destruction, is embedded in our future use of artificial intelligence if we do not pay attention to the data that feed the algorithms. The data collection and aggregation is the weakest link in the future of machine-supported decision-making.

Jan Kallberg, Ph.D.

Business leaders need to own cyber security

Consultants and IT staff often have more degrees of freedom than needed. Corporate cybersecurity requires a business leader to make the decisions, be personally invested, and lead the security work the same way as the business. The intent and guidance of the business leaders need to be visible. In reality, this is usually not the case. Business leaders rely on IT staff and security consultants to “protect us from cyberattacks.” The risk is obvious – IT staff and consultants are not running the business, lack complete understanding of the strategy and direction, and therefore are unable to prioritize the protection of the information assets.
Continue reading Business leaders need to own cyber security

Demilitarize civilian cyber defense

An cyber crimes specialist with the U.S. Department of Homeland Security, looks at the arms of a confiscated hard drive that he took apart. Once the hard drive is fixed, he will put it back together to extract evidence from it. (Josh Denmark/U.S. Defense Department)
U.S. Defense Department cyber units are incrementally becoming a part of the response to ransomware and system intrusions orchestrated from foreign soil. But diverting the military capabilities to augment national civilian cyber defense gaps is an unsustainable and strategically counterproductive policy.

The U.S. concept of cyber deterrence has failed repeatedly, which is especially visible in the blatant and aggressive SolarWinds hack where the assumed Russian intelligence services, as commonly attributed in the public discourse, established a presence in our digital bloodstream. According to the Cyberspace Solarium Commission, cyber deterrence is established by imposing high costs to exploit our systems. As seen from the Kremlin, the cost must be nothing because blatantly there is no deterrence; otherwise, the Russian intelligence services should have restrained from hacking into the Department of Homeland Security.

After the robust mitigation effort in response to the SolarWinds hack, waves of ransomware attacks have continued. In the last years, especially after Colonial Pipeline and JBS ransomware attacks, there has been an increasing political and public demand for a federal response. The demand is rational; the public and businesses pay taxes and expect protection against foreign attacks, but using military assets is not optimal.

Presidential Policy Directive 41, titled “United States Cyber Incident Coordination,” from 2016 establishes the DHS-led federal response to a significant cyber incident. There are three thrusts: asset response, threat response and intelligence support. Assets are operative cyber units assisting impacted entities to recover; threat response seeks to hold the perpetrators accountable; and intelligence support provides cyberthreat awareness.

The operative response — the assets — is dependent on defense resources. The majority of the operative cyber units reside within the Department of Defense, including the National Security Agency, as the cyber units of the FBI and the Secret Service are limited.

In reality, our national civilian cyber defense relies heavily on defense assets. So what started with someone in an office deciding to click on an email with ransomware, locking up the computer assets of the individual’s employer, has suddenly escalated to a national defense mission.

The core of cyber operations is a set of tactics, techniques and procedures, which creates capabilities to achieve objectives in or through cyberspace. Successful offensive cyberspace operations are dependent on surprise — the exploitation of a vulnerability that was unknown or unanticipated — leading to the desired objective.

The political scientist Kenneth N. Waltz stated that nuclear arms’ geopolitical power resides not in what you do but instead what you can do with these arms. Few nuclear deterrence analogies work in cyber, but Waltz’s does: As long as a potential adversary can not assess what the cyber forces can achieve in offensive cyber, uncertainties will restrain the potential adversary. Over time, the adversary’s restrained posture consolidates to an equilibrium: cyber deterrence contingent on secrecy. Cyber deterrence evaporates when a potential adversary understands, through reverse engineering or observation, our tactics, techniques and procedure.

By constantly flexing the military’s cyber muscles to defend the homeland from inbound criminal cyber activity, the public demand for a broad federal response to illegal cyber activity is satisfied. Still, over time, bit by bit, the potential adversary will understand our military’s offensive cyber operations’ tactics, techniques and procedures. Even worse, the adversary will understand what we can not do and then seek to operate in the cyber vacuum where we have no reach. Our blind spots become apparent.

Offensive cyber capabilities are supported by the operators’ ability to retain and acquire ever-evolving skills. The more time the military cyber force spends tracing criminal gangs and bitcoins or defending targeted civilian entities, the less time the cyber operators have to train for and support military operations to, hopefully, be able to deliver a strategic surprise to an adversary. Defending point-of-sales terminals from ransomware does not upkeep the competence to protect weapon systems from hostile cyberattacks.

Even if the Department of Defense diverts thousands of cyber personnel, it can not uphold a national cyber defense. U.S. gross domestic product is reaching $25 trillion; it is a target surface that requires more comprehensive solutions.

First and foremost, the shared burden to uphold the national cyber defense falls primarily on private businesses, states and local government, federal law enforcement, and DHS.

Second, even if DHS has many roles as a cyberthreat information clearinghouse and the lead agency at incidents, the department lacks a sizable operative component.

Third, establishing a DHS operative cyber unit is limited net cost due to higher military asset costs. When not engaged, the civilian unit can disseminate and train businesses as well as state and local governments to be a part of the national cyber defense.

Establishing a civilian federal asset response is necessary. The civilian response will replace the military cyber asset response, which returns to the military’s primary mission: defense. The move will safeguard military cyber capabilities and increase uncertainty for the adversary. Uncertainty translates to deterrence, leading to fewer significant cyber incidents. We can no longer surrender the initiative and be constantly reactive; it is a failed national strategy.

Jan Kallberg

Inflation – the hidden cyber security threat

 


Image: By Manuel Dohmen – Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=185802

In cyberspace, the focus is on threats from malicious activity — a tangible threat. A less obvious threat to cyber is inflation which undermines any cyber organization by eroding budget and employee compensation. Inflation can create unseen resignation rates if not addressed, and jeopardize ongoing cyber efforts and the U.S. Defense Department’s migration to cloud-based services. The competition for cloud security talent is razor-sharp in the private sector already.

There are different ways to build and maintain a cyber workforce: recruit, retrain and retain. The competition between the DoD and the private sector for talent will directly affect recruitment and retainment. Inflation and the shortage of skilled cyber professionals create increasing competition between the federal and private sectors for much-needed talent. Retraining professionals to become a part of the cyber workforce is costly, and if the incentives are not in place to stay in the force, it is short-lived as retrained cyber talent moves on. Inflation creates a negative outlook for recruiting, retraining, and retaining cyber talent.

The inflation expectations in 2022 are the highest in decades, which will directly impact the cost to attract and retain a cyber workforce. Even if the peak inflation is temporary due to COVID-19 as well as disruptions in the supply chain and the financial markets, the pressure on increased compensation is a reality today.

What does it mean in practical terms?

According to the Wall Street Journal, salaries will increase in 2022 for white-collar professionals in the range of 10%, and the federal workforce can expect an increase of less than a third of the gains in the private sector. These signs of growing salary gaps are likely far more severe and exacerbated in the cyber workforce.

For example, by browsing the current jobs ads, a manager for incident response in Rhode Island offers $150,000-$175,000 with the ability to work from home with zero commuting. A fair guess would be there’s a federal GS pay scale at 20-30% less, with work taking place from 8:30 a.m. to 4:30 p.m. in a federal facility; not to mention cloud security, where large players such as Amazon Web Services are actively recruiting from the federal sector.

An increasing salary gap directly impacts recruitment, where the flow of qualified applicants dries up due to the compensation advantage of the private sector. Based on earlier data, the difference in salary will trigger decisions to seek early retirement from the DoD, to pursue a second civilian career or to leave federal service for the private sector as a civilian employee.

The flipside of an all-volunteer force is that in the same way service members volunteer to serve, individuals have the option at the end of their obligation to seek other opportunities instead of reenlistment. The civilian workforce can leave at will when the incentives line up.

Therefore, if we face several years of high inflation, it should not be a surprise that there is a risk for an increased imbalance in incentives between the public and the private sectors that favor the private sector.

The U.S. economy has not seen high inflation since the 1970s and the early 1980s. In general, we all are inexperienced with dealing with steadily increasing costs and a delay of adjusted budgets. Inflation creates a punctured equilibrium for decision-makers and commanders that could force hard choices, such as downsizing, reorganization, and diluting the mission’s core goal due to an inability to deliver.

Money is easy to blame because it trespasses other more complex questions, such as the soft choices that support cyber talent’s job satisfaction, sense of respect, and recognition. It is unlikely that public service can compete with the private sector regarding compensation in the following years.

So to retain, it is essential to identify factors other than the compensation that make cyber talent leave and then mitigate these negative factors that lower the threshold for resignation.

Today’s popular phrase is “emotional intelligence.” It might be a buzzword, but if the DoD can’t compete with compensation, there needs to be a reason for cyber talent to apply and stay. In reality, inflation forces any organization that is not ready to outbid every competitor for talent to take a hard look at its employee relationships and what motivates its workforce to stay and be a part of the mission.

These choices might be difficult because they could force cultural changes in an organization. Whether dissatisfaction with bureaucracy, an unnecessary rigid structure, genuinely low interest for adaptive change, one-sided career paths that fit the employer but not the employee, or whatever reason that might encourage cyber talent to move on, it needs to be addressed.

In a large organization like the DoD and the supporting defense infrastructure, numerous leaders are already addressing the fact that talent competition is not only about compensation and building a broad, positive trajectory. Inflation intensifies the need to overhaul what attracts and retains cyber talent.

Jan Kallberg, Ph.D.