AI Digest
JUNE 2025
'AI and The Law'
authored by SAFE AI Foundation
~ ~ \\ // ~ ~
AI is a technology with wide impact, cutting across multiple industries. Not only does AI offers a great deal of automation but also intelligence! AI can learn fast and perform tasks faster and better than humans, and hence threatening human jobs. However, the usefulness of AI has prompted bad users to misuse this technology for a variety of crimes and frauds.
A list of AI abuses and misuses (Bad AI) include..
1. GenAI - violation of copyrights of artwork, music, images, video, etc.
2. Using ChatGPT to cheat in interviews, exams, etc.
3. Deepfakes – fake someone saying something they didn’t; impersonate
4. Fraud – wire and financial frauds by faking authorization for transactions, etc.
5. Digital Replicas – misuse and identity theft
6. Social media – fake news, flooding social media with inaccurate and biased messages, etc.
The scale of potential “chaos” cannot be under-estimated and hence proper “control” must be put in place to combat against malicious acts and intents.
How can such issues of AI misuse and abuses be addressed and contained?
The answer lies on the presence of both internal and external solutions needed to stop Bad AI!
1. Internal Solutions
(Part a) mostly refers to technology-oriented solutions created to stop bad users from misusing AI, for example, stop at the prompt, or violation checks before producing outputs for users. Other technological solutions include those that fights against AI-based attacks to systems, fight frauds, etc. These solutions are created by AI and technology companies.
(Part b) AI companies to practice strong AI principles and ethics in the development of AI software and products. This will close up any loopholes or weaknesses in the AI products, rendering malicious users unable to exploit these products for bad deeds. AI companies may also consider adopting AI management standards, such as ISO/IEC 42001. This standard helps companies to build trust, achieve AI compliance and align with best practices. It also ensures responsible AI development, deployment and operation. However, ISO/IEC 42001 requires certification and companies have to establish own measures and methods to meet certification requirements.
2. External Solution
These are mostly laws and regulations that prosecute bad users when they are caught, introduce fines or jail terms. These will deter others from committing such acts. In fact, several states have rolled out laws to protect against Bad AI, such as in California, Colorado, and Illinois (See references). So far, the US has not introduced a Federal AI Law. The EU, however, has introduced the EU AI Act, which is applicable to all European countries under its arm.
The Roles of Engineers and Legislators: Laws are needed to protect people of the State and Country while at the same time, provide ample opportunities for innovation and economic growth. This balancing act has to be performed by the state legislators, through consulting and working with a committee of experts, policy makers, and industry leaders. The law is not an area to work on by engineers. Engineers get to explain the AI technologies, how useful these technologies are and also highlight their shortcomings. Legislators have to look at the downside of new technologies and evaluate the possible harm and impact before deciding whether a new law is needed. Such a process requires considerable thought, consultation, voting and approval.
Society’s reaction to AI Rules and Laws: We need to look at how society will react when such laws are introduced. Do they welcome them? Do they obey them? Proper communication to the public is needed to ensure that the public understand the needs for these laws and gain their support in adhering to it. To date, the public seems to be receptive to new AI laws and are supportive on recognizing how AI can improve their lives. An emerging concern is jobs and authorities are now looking into how jobs displacement can be addressed.
In conclusion, just as early computer operating system crashes often, motivating for the improvement and creation of a new stable operating system and how computer viruses are handled through penalties, safeguard and protection, our society can be protected through a combination of technology and company solutions, state and government laws and regulations. By getting this right, our society can enjoy the benefits of AI without being fearful or hurt by it.
REFERENCES
[1] Safe AI Foundation – “AI Manifestation Unfolded” - https://safeaifoundation.com/vip-preview
[2] Colorado AI Act - https://www.skadden.com/insights/publications/2024/06/colorados-landmark-ai-act
[3] Illinois AI Act - https://www.morganlewis.com/pubs/2024/09/illinois-passes-new-law-to-address-ai-in-the-workplace
[4] California AI Act - https://www.pillsburylaw.com/en/news-and-insights/california-ai-laws.html
[5] EU AI Act - https://artificialintelligenceact.eu/
[6] ISO/IEC 42001 - https://www.iso.org/standard/81230.html
Disclaimer: The information in this digest is provided “as it is”, by the SAFE AI FOUNDATION, USA. The use of the information provided here is subject to the user’s own risk, accountability, and responsibility. The SAFE AI FOUNDATION is not responsible for the use of the information by the user or reader.
Note: The SAFE AI Foundation is a non-profit organization registered in the State of California and it welcomes inputs and feedback from readers and the public. If you have things to add concerning AI governance, acts, laws and regulations or would like to volunteer, please email us at: contact@safeaifoundation.com
