The balance between the limitations and benefits of artificial intelligence (AI) is complex and often misunderstood. AI provides considerable advantages, including improved efficiency through automation, informed decision-making with rapid data analysis, tailored user experiences, and the potential for business growth.
At the same time, AI suffers from significant disadvantages, including biases derived from outdated or contaminated training data, a lack of contextual understanding, ethical concerns related to privacy and data governance, and the risk of job displacement. A thorough comprehension of this intricate relationship is crucial to guarantee responsible AI implementation and use.
Successful organizations prioritize ethical practices, continuously monitoring to identify and remove biases and ensuring decision-making processes and regular audits have human oversight. By promoting transparency and understanding AI’s strengths and constraints, organizations can effectively harness its advantages while minimizing risks, resulting in more ethical and responsible use.
Businesses Turn To AI for Productivity, Efficiency, and Growth
AI presents a unique opportunity for companies with the potential for innovation, boosted productivity, and growth. McKinsey research estimates that GenAI could add up to $4.4 trillion in economic value to the global economy. For example, Amazon has increased efficiency by using AI to streamline delivery routes, create more intelligent warehouse robots, and better predict where to stock new items. Google uses AI systems to generate more than 25 percent of new product code. Despite the reliance on AI, Google still depends on human programmers to review the code and make high-level decisions.
In the healthcare sector, AI can streamline the exchange of documents and records through electronic data interchange (EDI) systems. EDI encompasses the communication protocols, electronic processes, and international standards that ensure data is exchanged accurately and efficiently, which is particularly useful in highly regulated fields like healthcare. Healthcare EDI software helps to encrypt and transfer protected patient data and transmit details, including coordination of benefits, claims status and information, insurance data, payment data, and eligibility forms. EDI helps improve data accuracy, operational efficiency, and patient care by monitoring patient health and tracking outcomes.
AI reduces human errors in various ways, such as flagging potential mistakes before they occur or ensuring consistency and precision in medical applications. For example, AI-guided surgical robotics, like the da Vinci robot, allows a human surgeon to perform all surgical gestures in a console setting. The da Vinci robot cannot act independently without human input.
AI Poses Ethical Dilemmas
While AI boasts a plethora of benefits in business and life, it’s not without possible pitfalls and ethical concerns. Algorithmic bias occurs when algorithms make decisions that systematically disadvantage certain groups of people. This happens because those algorithms may have been trained on unrepresentative or outright racist data. This bias can have disastrous results when used to screen job candidates. For example, in 2018, Amazon scrapped an AI-based recruiting tool that “taught itself that male candidates were preferable.”
This unintentional bias led to a discriminatory hiring practice. In another case, the U.S. Equal Employment Opportunity Commission (EEOC) alleged that tutoring company iTutor programmed its software to automatically reject female applicants over 55 and male applicants over 60. This bias resulted in iTutor’s failure to hire over 200 qualified older applicants from the United States because of their age. As Hilke Schellmann, author of “How AI Can Hijack Your Career and Steal Your Future,” said in a BBC interview, “One biased human hiring manager can harm a lot of people in a year, and that’s not great. But an algorithm that is maybe used in all incoming applications at a large company…that could harm hundreds of thousands of applicants.”
Cybersecurity Risks and the Largest Healthcare Data Breach in History
AI poses other significant security and privacy risks, including data breaches, security vulnerabilities, and the misuse of AI by cybercriminals. These attacks are not limited to large corporations or government entities; any industry is vulnerable to malicious cyberattacks, and the ramifications can be catastrophic.
For example, in 2024, Change Healthcare, a UnitedHealthcare Group (UHG) subsidiary, became the victim of a massive ransomware attack in which foreign cybercriminals stole Health Insurance Portability and Accountability Act ( HIPAA)-protected medical information, encrypted company files, and demanded money to unlock them. The attack impacted claims processing and payments, preventing patients from getting care and prescriptions. Rick Pollack, President and CEO of the American Hospital Association, called the cyberattack “the most significant and consequential incident of its kind against the U.S. healthcare system in history.”
Patient information leaked in the breach, including passwords, medical treatment received, billing addresses, and credit card numbers, have since appeared on the dark web. Axios reported that UHG leaders estimate the attack could cost the company $2.45 billion. In March of 2024, the Department of Health and Human Services Office for Civil Rights (OCR) investigated whether Change Healthcare had been HIPAA-compliant during the attacks. The HIPAA Journal reported the breach affected the personal information of 100 million people (nearly one in three individuals), making it the worst healthcare data breach in the country. Indeed, the full scope of ramifications is yet unknown.
Human Error and Cybersecurity
Companies can invest in cybersecurity strategies and tools to mitigate cybersecurity threats and protect consumer data, especially since 74 percent of data breaches are attributed to human error. According to Infosec, nearly three out of four incidents involved a human element, such as privilege misuse, stolen credentials, or social engineering. For example, if an employee sends sensitive data to a personal computer or staff members try to access a system they don’t have authorization to use, they could unknowingly open a system vulnerability for an attack.
Locking Into Security Solutions
The number one way to protect an organization is to educate employees on the cyber threats they might encounter. Cybersecurity awareness training helps prevent innocent errors and assists staff members in being more proactive. Implementing AI tools to identify and flag employee risks can help a company or healthcare system educate its workforce on which data is sensitive and how to work with that data.
A well-trained and competent cybersecurity team can help organizations protect their customers and assets with technical defense tools such as network and web application firewalls, intrusion detection systems (IDS), demilitarized zones (DMZs), and security information and event management (SIEM). Additional tools include data loss prevention (DLP), patch management, and honeypots.
How to Use AI Responsibly
Much has been written about AI’s benefits and drawbacks. While AI has numerous productivity and analytical benefits, it’s imperative for organizations to be aware of the ethical and security problems and ensure that robust security practices and data privacy policies are in place to combat potential risks. This includes appropriately vetting AI tools for business needs and risks before purchasing, employing human oversight, and regularly auditing AI systems. It’s also essential to ensure that humans, not AI, are engaged in decision-making regarding patient care, especially for vulnerable populations.
As technology changes, the human approach to AI must change as well. Understanding the organization’s context and how AI will impact employees’ and users’ day-to-day operations, security, and privacy is vital to finding the right AI tools for a business. Adopting a robust framework with best practices and safeguards will be critical to ensuring data security and privacy for all.
