AI Crisis Management: Strategic Planning

Atlas Marketing.Business.Crisis Communications.Digital Marketing.Marketing Strategy
AI Crisis Management: Strategic Planning

Artificial Intelligence – or AI as it’s more commonly referred to – is not only the hot new buzzword in the business world today but is also quickly becoming a business advantage for early adopters. This technology impacts industries such as finance, marketing, healthcare, manufacturing, and construction creating productivity efficiencies. However, many companies overlook the crisis management implications of AI adoption and the impact the tool may have when issues arise.

With any new technology, assessing its impact, reach, and efficacy is vital to continued adoption. Even as AI is still in its commercial infancy, there have been several crises that have potentially blemished the future of the popular tech.

Examples include Microsoft’s Tay Chatbot – within 24 hours of launch, the bot posted offensive and inappropriate topics forcing the company to take the technology offline; Amazon’s internal recruitment tool – after installing an AI-based recruiting tool, the company found that the bot was biased against female applicants and shut the tool down; and Apple’s Face ID – upon launching the security feature the media reported that people with darker skin tones were having challenges getting past the technology.

There are several other notable examples, but the business ramifications remain massive in each. These specific crises can cost upwards of hundreds of millions of dollars not only in anticipated commercial gain, but research and development, technological infrastructure, and marketing costs. With any crisis, the need to understand the potential impact is vital to successfully managing the situation.

Proactive Crisis Planning

AI has significant upsides, but one lingering issue is that companies who integrate the software without having a plan in place can quickly find themselves in a crisis due to misuse of the technology. In my experience with crisis management, I’ve found that approximately 60% of situations, on average, can be planned for. While this might seem like a favorable margin, this leaves a massive opportunity for unplanned incidents to occur, grow, and quickly overwhelm at the most inopportune time. Even companies who plan and actively practice crisis response protocols can be overcome by an unplanned event. Given that AI is a generative technology that operates in the snap of a finger, one can imagine how that gap in planning can be exploited, and quickly.

A standard crisis plan should first contain a list of potential crises and situations along with the team leads and members who will manage, communicate, and stay informed of the event in all stages. Within company plans, standard best practice is to limit those involved in crisis response. However, with technology at the heart of a crisis, the impact radius becomes larger, more impactful, and includes many outside counselors. For example, there may be a need to consult with IT, legal, PR, or insurance, along with the C-Suite, before any responses are generated. It is important to know who those professionals are and begin building a relationship immediately.

There are multiple components to an AI crisis plan to consider in addition to the foundational elements of a crisis management plan:

  • Risk assessment: identify vulnerabilities within a system from a development standpoint, integration, user perspective, and commercial aspect.
  • Reach: how many users will be impacted by an emergency and what is the geographic reach?
  • Client interaction: what impact will a crisis have on trust and credibility in existing or future business relationships?  
  • Ease of shutdown: is this tool entwined in other software?

How to Prepare – Best Practices

In addition to plan development best practices, there are several items that will strengthen a company’s AI-based crisis response.

Regular review and updates of IT crisis plans are critical as the technology continues to evolve. Moore’s Law states that every two years computational power and progress doubles. This requires companies to maintain a status review on a regular basis.

It’s important to build a team of trusted AI and tech experts to ensure the company is up to speed on emerging technology, trends, and threats. Companies may even consider attending a conference or subscribing to various newsletters as a method of surveying the industry and learning about the multiple options available. For reference, we’ve created a list of top conferences and newsletters for your consideration.

Companies should consider using the latest technology in crisis detection and management. At its core, AI combines advanced analytics, real-time monitoring, and algorithms, which can help to pinpoint incoming or potential situations. New programs are emerging within the market often, so be sure to do your research on what tech packages would work best for you.

An often overlooked step, it’s critical to create a company-wide AI policy that details use and defines access. Atlas Marketing established a policy and works toward transparency in use. Our team will notify clients when we use the tool, our steps to ensure security, and ensures a review of all generative content. This policy strengthens our decision-making process, acknowledges the tool’s limitations and strengths, and helps build trust with our partners and clients.

AI and the supporting technology continue to evolve. With that evolution, there are many reasons to remain focused on crisis management as an aspect of a company’s AI approach.

Cybersecurity remains one of the leading business challenges. In fact, only 4% of companies recently surveyed feel confident in their ability to protect their devices from an attack. Adding this component to your crisis management plan is essential as, according to a University of Maryland study, a cyber-attack occurs every 39 seconds. These two statistics alone show the importance of expanding your definition of a crisis to include technology, AI, and cybersecurity.

Companies must consider the ethical component of AI as well in their planning. The potential for misinformation via the tool and a lack of human oversight adds complexity to the planning process that most other crises do not. The inclusion of a stated policy regarding human oversight and review is integral to the acceptance of many insurance policies as well. Atlas Marketing anticipates that many companies will find this to become of larger significance in the insurance renewal process adding to the complexity as well as time investment.

AI Crisis Management Takeaways

Safeguarding operations, reputation, and customer trust will not only strengthen a company’s profits, but those who plan accordingly will mitigate the multitude of risks involved with utilizing AI to its fullest capacity.

As technology pushes industries to improve productivity and efficiencies, companies that understand the various risks and develop crisis management plans that address the anticipated technological challenges will be in the best position for success.

Research for this article included the use of Artificial Intelligence.