Fact Check: Is artificial intelligence trustworthy

Fact Check: Is Artificial Intelligence Trustworthy?

What We Know

The question of whether artificial intelligence (AI) is trustworthy has garnered significant attention in recent years. According to the 2025 AI Index Report, AI's influence on society is profound, with advancements leading to increased applications in various fields such as healthcare and transportation. However, the report also highlights that while AI systems are becoming more capable, the public's trust in these technologies remains a critical challenge. A global study indicated that more than half of people worldwide are unwilling to trust AI, reflecting widespread skepticism about its reliability and safety (Global study reveals trust of AI remains a critical challenge).

Experts from UC San Diego emphasize that while AI can enhance human capabilities, it is essential to approach its trustworthiness with caution. David Danks, a professor of data science, notes that many AI systems should not be trusted at this stage due to their inherent limitations and potential for error (How Trustworthy is AI?). Furthermore, the National Institute of Standards and Technology (NIST) has published guidelines stressing the importance of responsible AI development, which includes addressing adversarial risks and ensuring transparency in AI operations (NIST Trustworthy and Responsible AI Report).

Analysis

The evidence surrounding AI's trustworthiness presents a mixed picture. On one hand, the advancements in AI technology are impressive, with significant improvements in performance metrics and applications in real-world scenarios. For instance, the AI Index Report notes that AI systems have made substantial strides in generating high-quality outputs and outperforming humans in specific tasks (2025 AI Index Report). This progress suggests that AI can be reliable in certain contexts.

However, the concerns raised by experts and studies cannot be overlooked. The global study on trust in AI reveals that public skepticism is rooted in the opacity and unpredictability of AI systems. Many users are willing to experiment with AI but remain hesitant to rely on it for critical decisions (Global study reveals trust of AI remains a critical challenge, How Trustworthy is AI?). The complexity of AI algorithms often leads to unexpected errors, which can undermine trust. Moreover, the NIST report highlights the need for robust frameworks to ensure that AI systems are not only effective but also safe and transparent (NIST Trustworthy and Responsible AI Report).

The sources consulted provide a balanced view, with credible institutions like Stanford HAI and NIST contributing to the discourse on AI's trustworthiness. However, the ongoing evolution of AI technology means that both its capabilities and the public's perceptions are likely to change over time.

Conclusion

The claim that artificial intelligence is trustworthy is Partially True. While AI systems are becoming increasingly capable and are being integrated into various sectors, significant concerns regarding their reliability, transparency, and the public's trust remain. The evidence indicates that while AI can perform well in specific applications, it is essential to approach its use with caution and to continue developing frameworks that enhance its trustworthiness.

Sources

1. The 2025 AI Index Report | Stanford HAI
2. Global study reveals trust of AI remains a critical challenge
3. How Trustworthy is AI?
4. NIST Trustworthy and Responsible AI Report Adversarial Machine Learning
5. Trust, attitudes and use of Artificial Intelligence: A global study
6. A call for transdisciplinary trust research in the artificial intelligence domain
7. Trust, attitudes and use of artificial intelligence: A global study
8. Trustworthiness of Artificial Intelligence - IEEE Xplore