In an age where anything that can be sensed is recorded, it’s becoming increasingly untenable for government and military organizations to rely solely on human processing power alone. Defense analysts are expected to keep their eyes on more people and places than ever before, and our nation’s security posture is growing increasingly complex – so much so that there is simply too much data for them to bring together and analyze in the short time frames required for mission success. Agencies rely on the wisdom and expertise of their analysts and that will never change, but without the computing capabilities of machines, our nation’s readiness will degrade as analysis fails to keep pace with incoming data and the expanding needs of the military.
The benefits of artificial intelligence (AI) technologies are becoming more universally evident and accepted. Not only can these solutions save analysts time and allow them to devote more of their energy to higher-level analysis, they also make it easier for those in charge to make informed decisions. The potential value of these solutions is driving leaders in the defense community to push for AI’s widespread adoption. The Department of Defense, for example, recently ordered the creation of the Joint Artificial Intelligence Center (JAIC), with the goal of quickly launching a series of AI projects, and accelerating the implementation of these new technologies across all department missions. While this is an encouraging step, successfully integrating AI and other analytic tools into current analyst workflows remains an uphill battle.
The challenges with current approaches can all be traced to the same root cause: in the rush to bring AI and other technologies to intelligence missions, the analyst has been largely left out of the equation. The impulse has been to develop technologies first, and then figure out how to deploy them. Analysts may not accept and use the AI-informed analytics, either because they don’t trust the outputs, or because they fear the computers will put them out of a job. The technologies introduced by data scientists and computer engineers are generally created without the input and buy-in of the analysts. This often means the AI that is introduced is intricate and not user friendly. Worse, when data scientists get too caught up in what the technology can do on an academic and theoretical level, the analytic tools may lack the necessary context to be truly relevant to the specific mission, and ultimately cannot be relied upon to support decision-making.
The relationship between humans and machines is often presented as an adversarial one: success or advancement of one is often associated with the degradation of the other. But that’s not actually the case. Humans and machines can work together as partners and strengthen each other in kind.
If government and military missions want analysts to embrace these new technologies and apply them effectively in the long-term, they’ll need to adopt a new paradigm: “Analyst 2.0.” This means letting machines do what they do best so that people can do what they do best. Rather than leaving the analysts out, organizations ought to make them central to every aspect of developing and deploying AI – thus positioning analysts for success and ensuring that the tools are tailored to an analyst’s operational environment and mission. When analysts play a key role in bringing analytics to the mission, the outputs are much more likely to be accurate and contextualized to the mission. The tools are more likely to be transparent, accessible and trusted; and the analysts themselves can more clearly see the value of their changing role.
No AI technology on the horizon can replace human judgment. Technology is an important piece of Analyst 2.0, but it alone will not enhance national security. By making sure new intelligence tools are not just AI-informed, but analyst-informed, organizations can tap the potential of advanced analytics to empower analysts and enhance operational and mission effectiveness.
No comments:
Post a Comment