Artificial intelligence tools have become part of everyday life. From AI chatbots that assist with work tasks to AI “companions” designed for personal conversations and emotional support, these systems are increasingly sophisticated and increasingly common. You only have to look at some recent movies like “Her” or “The Creator” to understand how AI issues may eventually create security concerns.
For security clearance holders and applicants these technologies raise a number of important security concerns that are often overlooked. It has caused me to reconsider how these issues could potentially affect my clients.
After 27 years of representing security clearance holders and applicants, I have seen how emerging technologies frequently outpace public understanding of the security risks involved. AI companions and chatbot platforms present unique issues involving privacy, disclosure, judgment, foreign influence, and handling of sensitive information.
The Privacy and Data Collection Problem
Many AI chatbot and companion platforms collect extensive user data, including:
- Personal conversations
- Behavioral information
- Location and device data
- Preferences and emotional interactions
In some cases, users disclose highly personal or sensitive information without fully understanding how the data is stored, processed, or shared. It can be easy to ask questions or provide very personal information to these platforms and assume it stays with the chatbot or companion.
For clearance holders, this creates potential Guideline M, Information Technology concerns about:
- Vulnerability to exploitation
- Exposure of sensitive personal information
- Improper handling of government or work-related material
The risk becomes even greater if the platform is owned, hosted, or supported by entities tied to foreign countries. Many foreign governments have invested in such programs.
Sharing Sensitive or Work-Related Information
One of the most significant risks involves users treating AI tools like trusted assistants and entering information that should never be disclosed.
Even seemingly harmless prompts may involve:
- Internal government processes
- Sensitive work discussions
- Procurement or technical information
- Operational details
In classified or cleared environments, entering protected or sensitive information into unauthorized AI systems can create security clearance, employment and even potential criminal issues.
Foreign Ownership and Influence Concerns
Some AI platforms and applications also have foreign ownership, foreign data hosting, or ties to companies operating in countries considered adversarial to the United States.
Under the adjudicative guidelines, Security Executive Agent Directive 4 (SEAD 4), clearance holders are expected to exercise sound judgment regarding foreign influence and data security risks. Use of platforms with unclear ownership or foreign ties may invite additional scrutiny, particularly if sensitive information is involved.
As with social media and communication platforms before them, AI applications are likely to receive increasing attention from agencies and security offices. We are already seeing questions about usage in our Intelligence Community security clearance cases.
AI Companions and Personal Conduct Concerns
AI companion applications raise an additional layer of concern because they are specifically designed to create ongoing personal interaction and emotional engagement.
While using these applications is not inherently disqualifying, excessive reliance on them or disclosing highly sensitive personal information through them could potentially raise questions regarding judgment, reliability, or susceptibility to manipulation. One example, as portrayed in the movie “Hers” would be in having a significant friendship or romantic relationship with an AI companion that could be the basis of coercion by a foreign power.
The concern is not AI companions themselves, but how the individual uses it.
A Potential Example
Consider a clearance holder who regularly uses an AI chatbot to help organize work tasks and draft summaries. Over time, the individual begins entering increasingly detailed work-related information into the system, believing the material is unclassified and harmless.
Later, an agency determines that some of the information should not have been entered into an unauthorized external platform. The issue then becomes not only the disclosure itself, but also whether the individual exercised appropriate judgment in handling potentially sensitive information.
Policies and Laws Are Still Evolving
One challenge in this area is that agency policies regarding AI use are evolving rapidly. Some agencies permit limited use of approved AI tools, while others prohibit certain platforms entirely. Another challenge is that many federal employees and contractors are not always aware of changing AI policies. An Executive Order has been issued and other congressional legislation involving AI is being considered.
This creates confusion for many clearance holders who may assume that publicly available AI tools are automatically permissible. In reality, approval often depends on:
- The specific platform being used
- The type of information involved
- Agency-specific guidance and restrictions
What is acceptable in one environment may create serious issues in another.
Final Thoughts
AI companions and chatbot technologies are becoming increasingly integrated into daily life, but security clearance holders should approach these tools cautiously.
Understanding how these technologies may be viewed through the lens of the adjudicative guidelines is essential to protecting both a clearance and a career.
This article is for informational purposes only and should not be construed as legal advice. Security clearance rules, agency AI policies, and federal guidance may change, and readers should consult counsel regarding their specific circumstances.
