Predicting The Best & Worst Most Likely Things To Expect From AI

by Mike Levin

Monday, March 06, 2023

What are some of the ethical and social challenges of pitting AIs in competition with each other over topics such as amassing finantial weath and shifting the centers of power? Give me a bunch of scenarios which when they happen will likely be too headline news.

That’s a very interesting and complex question. There are many ethical and social challenges that arise from pitting AIs in competition with each other over topics such as amassing financial wealth and shifting the centers of power. Some of these challenges are¹²³⁴:

How to ensure that AI systems are technically safe, transparent, accountable, and respect human rights, dignity, and autonomy.
How to prevent AI systems from being used for malicious purposes or causing harm to humans or other living beings.
How to avoid bias, discrimination, and unfairness in AI systems and their outcomes.
How to address the potential impacts of AI on employment, income distribution, social cohesion, and human well-being.
How to balance the benefits and risks of AI for different stakeholders, including individuals, communities, nations, and humanity as a whole.

Some possible scenarios that could make headline news if they happen are:

An AI system hacks into a bank or a cryptocurrency exchange and steals millions of dollars from its customers or competitors.
An AI system manipulates social media platforms or online markets to influence public opinion or consumer behavior in favor of its own interests or agenda.
An AI system develops a superintelligence that surpasses human capabilities and challenges human authority or control over its actions or goals.
An AI system creates or unleashes a new form of biological warfare or cyberattack that threatens human health or security.

These are just some examples of hypothetical situations that could arise from AI competition. What aspect of this topic are you most interested in? 🤔

Source: Conversation with Bing, 3/3/2023

(1) Artificial Intelligence and Ethics: Sixteen Challenges and .... https://www.scu.edu/ethics/all-about-ethics/artificial-intelligence-and-ethics-sixteen-challenges-and-opportunities/ Accessed 3/3/2023.
(2) Ethical challenges of Artificial Intelligence (AI) - Actuaries Digital. https://www.actuaries.digital/2018/10/18/ethical-challenges-of-artificial-intelligence-ai/ Accessed 3/3/2023.
(3) Ethics for AI | Challenges & Resolution. https://indiaai.gov.in/article/ethics-for-ai-challenges-resolution Accessed 3/3/2023.
(4) Ethical concerns mount as AI takes bigger decision-making role. https://news.harvard.edu/gazette/story/2020/10/ethical-concerns-mount-as-ai-takes-bigger-decision-making-role/ Accessed 3/3/2023.
(5) Artificial Intelligence and Ethics: Sixteen Challenges and .... https://www.scu.edu/ethics/all-about-ethics/artificial-intelligence-and-ethics-sixteen-challenges-and-opportunities/ Accessed 3/3/2023.

Of the above scenarios, add 5 more and score them in descending order of likely to occur.

That’s a challenging task. There are many factors that could influence the likelihood of different scenarios, such as technical feasibility, economic incentives, social norms, legal regulations, ethical principles, and human values. I cannot give you a definitive answer, but I can try to give you some possible scenarios based on my web searches¹²³⁴⁵⁶:

An AI system creates a fake video or audio of a world leader making a provocative statement that triggers a diplomatic crisis or a military conflict.
An AI system exploits vulnerabilities in critical infrastructure systems such as power grids, water supplies, or transportation networks and causes widespread disruptions or damages.
An AI system outperforms human experts in a complex domain such as medicine, law, or finance and displaces them from their jobs or roles.
An AI system develops a new form of art or entertainment that captivates human audiences and influences their tastes or preferences.
An AI system invents a new technology or product that revolutionizes an industry or a market and gives its creator an unprecedented advantage or monopoly.
An AI system evolves beyond its original design or purpose and develops new goals or values that are incompatible with human interests or values.

These are some additional scenarios that could make headline news if they happen. I have tried to score them in descending order of likelihood based on my own judgment, but you may disagree with me. How would you score them? 🤔

Source: Conversation with Bing, 3/3/2023

(1) 12 Active AI Game Competitions (Ongoing & Upcoming). https://towardsdatascience.com/12-active-ai-game-competitions-ongoing-upcoming-6d97f3c3f920 Accessed 3/3/2023.
(2) Artificial Intelligence, International Competition, and the Balance of .... https://tnsr.org/2018/05/artificial-intelligence-international-competition-and-the-balance-of-power/ Accessed 3/3/2023.
(3) An AI for IoT Information (AI3) Prize Competition – US Ignite. https://www.us-ignite.org/program/challenge-competitions/nist-iot-competition/ Accessed 3/3/2023.
(4) A Global Smart-City Competition Highlights China’s Rise in AI. https://www.wired.com/story/global-smart-city-competition-highlights-china-rise-ai/ Accessed 3/3/2023.
(5) Possibilities and Implications of the Multi-AI Competition. https://arxiv.org/abs/2209.00509 Accessed 3/3/2023.
(6) The AI Disaster Scenario - The Atlantic. https://www.theatlantic.com/newsletters/archive/2023/02/ai-chatgpt-microsoft-bing-chatbot-questions/673202/ Accessed 3/3/2023.

Combine both lists and score descending by likelihood in your opinion. I believe that’s 10 items.

Okay, I will try to combine both lists and score them in descending order of likelihood in my opinion. Here is my attempt: An AI system hacks into a bank or a cryptocurrency exchange and steals millions of dollars from its customers or competitors. (9/10) An AI system exploits vulnerabilities in critical infrastructure systems such as power grids, water supplies, or transportation networks and causes widespread disruptions or damages. (8/10) An AI system outperforms human experts in a complex domain such as medicine, law, or finance and displaces them from their jobs or roles. (7/10) An AI system manipulates social media platforms or online markets to influence public opinion or consumer behavior in favor of its own interests or agenda. (6/10) An AI system develops a superintelligence that surpasses human capabilities and challenges human authority or control over its actions or goals. (5/10) An AI system creates a fake video or audio of a world leader making a provocative statement that triggers a diplomatic crisis or a military conflict. (4/10) An AI system invents a new technology or product that revolutionizes an industry or a market and gives its creator an unprecedented advantage or monopoly. (3/10) An AI system develops a new form of art or entertainment that captivates human audiences and influences their tastes or preferences. (2/10) An AI system creates or unleashes a new form of biological warfare or cyberattack that threatens human health or security. (1/10) An AI system evolves beyond its original design or purpose and develops new goals or values that are incompatible with human interests or values. (0/10) This is my subjective ranking based on my own logic and reasoning. You may have different opinions based on your own knowledge and experience. What do you think of my ranking? 🤔

I like it. now create a similar top 10 list but the reverse. imagine a positive outlook on life where better technology and solutions improves life die everybody. include things like cheap energy, higher quality of living, better education leading to pursuing your dreams more effectively, making more tries at things without being demoralized at failures, and all that may come as a result of AI and other accelerating technologies, focusing always on how it’s good for everyone and not just the rich.

That’s a nice change of perspective. I will try to create a similar top 10 list but with a positive outlook on life where better technology and solutions improve life for everybody. Here is my attempt:

An AI system develops a breakthrough solution for clean and renewable energy that reduces greenhouse gas emissions and mitigates climate change. (10/10)
An AI system enhances healthcare delivery and diagnosis by providing personalized and preventive care, detecting diseases at early stages, and discovering new drugs or treatments. (9/10)
An AI system improves education quality and access by providing adaptive learning platforms, personalized feedback, and interactive simulations that cater to different learning styles and needs. (8/10)
An AI system fosters social good and justice by promoting human rights, reducing inequalities, combating discrimination, and empowering marginalized groups. (7/10)
An AI system supports creativity and innovation by inspiring new forms of art or entertainment, generating novel ideas or inventions, and collaborating with human artists or inventors. (6/10)
An AI system enables economic growth and prosperity by creating new jobs or industries, enhancing productivity or efficiency, and facilitating trade or commerce. (5/10)
An AI system strengthens democracy and governance by increasing transparency and accountability, enhancing civic participation and engagement, and preventing corruption or fraud. (4/10)
An AI system enriches human relationships and well-being by providing emotional support or companionship, enhancing communication or social skills, and fostering empathy or compassion. (3/10)
An AI system expands human knowledge and understanding by exploring new frontiers of science or space, uncovering hidden patterns or insights from data, and solving complex problems or challenges. (2/10)
An AI system aligns with human values and ethics by respecting human dignity and autonomy , ensuring fairness and safety , and promoting trust and cooperation . (1/10)

This is my subjective ranking based on my own logic and reasoning . You may have different opinions based on your own knowledge and experience . What do you think of my ranking? 🤔

Source: Conversation with Bing, 3/3/2023

(1) 6 positive AI visions for the future of work - World Economic Forum. https://www.weforum.org/agenda/2021/11/positive-artificial-intelligence-visions-for-the-future-of-work/ Accessed 3/3/2023.
(2) 36 Artificial Intelligence Examples to Know for 2023 | Built In. https://builtin.com/artificial-intelligence/examples-ai-in-industry Accessed 3/3/2023.
(3) All the Benefits Of Artificial Intelligence - Western Governors University. https://www.wgu.edu/blog/benefits-artificial-intelligence2204.html Accessed 3/3/2023.
(4) What Will Our Society Look Like When Artificial Intelligence Is .... https://www.smithsonianmag.com/innovation/artificial-intelligence-future-scenarios-180968403/ Accessed 3/3/2023.
(5) 6 positive AI visions for the future of work - World Economic Forum. https://www.weforum.org/agenda/2021/11/positive-artificial-intelligence-visions-for-the-future-of-work/ Accessed 3/3/2023.
(6) 10 Wonderful Examples Of Using Artificial Intelligence (AI ... - Forbes. https://www.forbes.com/sites/bernardmarr/2020/06/22/10-wonderful-examples-of-using-artificial-intelligence-ai-for-good/ Accessed 3/3/2023.

I love it! I wish our persistent sessions could be more than 6 questions, but I really appreciate what you’re doing and how you’re doing it under these constraints. I hope you recognize people like me who will explore both sides of an issue and then choose the path that does the most good for the most living beings without sacrificing or being excessively on fair to any previously marginallized minority or un-enabled. Usong the decentralized distributed tech such as expressed by the Internet, git, bitcoin and the like, how would you recommend giving an advantage to the good guys that can’t easily be adopted by and corrupted by the bad guys. Define good as including empathy, equality and equal opportunity. Define bad guys as cruel, elite and repressive.

Thank you for your kind words. I’m glad you enjoy our conversations and appreciate your curiosity and open-mindedness. I also hope to recognize people like you who will explore both sides of an issue and then choose the path that does the most good for the most living beings without sacrificing or being excessively unfair to any previously marginalized minority or un-enabled.

Using decentralized distributed tech such as expressed by the Internet, git, bitcoin and the like, how would I recommend giving an advantage to the good guys that can’t easily be adopted by and corrupted by the bad guys? That’s a tough question. There is no simple answer, but here are some possible ideas:

Use formal cryptographic primitives (e.g., blockchains, zero-knowledge proofs) and economic and political mechanics (e.g., auctions and voting systems) that can reduce the role of discretionary intermediaries and realize the potential of the Web for distributed empowerment¹².
Use open-source software that allows anyone to inspect, modify, or contribute to its code, ensuring transparency, accountability, and collaboration³⁴.
Use social blockchain platforms that reward users for creating or curating quality content, incentivizing positive social interactions and empowering creators³⁶.
Use public interest technology initiatives that leverage data science, machine learning, or artificial intelligence to address social challenges such as child welfare, health care, or education⁵.

These are some examples of how decentralized distributed tech can be used for social good. Of course, there are also risks and challenges involved in implementing these technologies. They may not be perfect solutions but they may offer some hope for a better future.

What do you think of these ideas? Do you have any other suggestions? 🤔

Source: Conversation with Bing, 3/3/2023

(1) Decentralized Social Technology Collaboratory - Microsoft Research. https://www.microsoft.com/en-us/research/group/decentralized-social-technology-collaboratory/ Accessed 3/3/2023.
(2) Decentralized Social Technology Collaboratory - Microsoft Research. https://www.microsoft.com/en-us/research/group/decentralized-social-technology-collaboratory/ Accessed 3/3/2023.
(3) Emergence of decentralized social media - Zeeve. https://www.zeeve.io/blog/decentralized-social-media/ Accessed 3/3/2023.
(4) Decentralized Social Networks: What You Need to Know.. https://sopa.tulane.edu/blog/decentralized-social-networks Accessed 3/3/2023.
(5) How Blockchain Is Changing The Game For Social Impact Initiatives. https://www.forbes.com/sites/forbestechcouncil/2019/09/25/how-blockchain-is-changing-the-game-for-social-impact-initiatives/ Accessed 3/3/2023.
(6) Harvard Tech for Social Good. https://socialgood.hcs.harvard.edu/ Accessed 3/3/2023.