To AI or not to AI?

It was the Fall of my first year at the University of Pennsylvania. It was the first day of a class. The title of the class: Programming Languages and Techniques (then CSE 120, now CIS 1200). Our professor reviewed the syllabus, introduced his teaching assistants, and finished describing the class project: ?write a LISP (https://en.wikipedia.org/wiki/Lisp_(programming_language) ) Interpreter in Pascal (https://en.wikipedia.org/wiki/Pascal_(programming_language) ). Say what? Pascal? LISP? Interpreter? Argh! For sure, I was going to get my money’s worth in this class. I did. ??

Why start this blog with that itty-bitty moment in time?. Simply, it represented my first foray into the world of artificial intelligence (AI) – more specifically, natural language processing (NLP). Not many years later, I stumbled on to what would become my favorite programming language of all time – Prolog (https://en.wikipedia.org/wiki/Prolog ) – another declared AI language. In addition to Pascal, LISP and Prolog, I ultimately learned to program in Assembler, COBOL, FORTRAN, PL1, and C. Why Prolog as my favorite – deserves another blog posting for another day – ha ha.

So here we are a few decades later, and the world of AI has come a long way. As AI has escaped the lab and academia into government and commercial markets, the use cases continue to explode. Here’s an interesting result from a query (don’t criticize my prompt – ha ha, I just typed) using ChatBox AI (GPT-3.5) on my mobile phone:

Me:

As a researcher, I wonder what the top 20 or so use cases that AI can help with?

ChatBox:

1.?????? Healthcare: AI can assist in diagnosing diseases, analyzing medical images, predicting patient outcomes, and personalizing treatment plans.

2.?????? Finance: AI can be used for fraud detection, risk assessment, investment strategies, customer service, and personalized financial advice.

3.?????? Retail: AI can help with inventory management, personalized recommendations, customer service, demand forecasting, and supply chain optimization.

4.?????? Marketing: AI can assist in targeted advertising, customer segmentation, sentiment analysis, and content optimization.

5.?????? Customer service: AI can be used for chatbots, virtual assistants, sentiment analysis, and customer feedback analysis.

6.?????? Education: AI can help with personalized learning plans, student assessment, automated grading, and educational content creation.

7.?????? Transportation: AI can assist in route optimization, traffic management, predictive maintenance, autonomous vehicles, and logistics.

8.?????? Manufacturing: AI can be used for predictive maintenance, quality control, production optimization, and supply chain management.

9.?????? Agriculture: AI can help with crop monitoring, yield prediction, pest detection, and irrigation management.

10.?? Energy: AI can assist in predictive maintenance, energy consumption optimization, grid management, and renewable energy integration.

11.?? Cybersecurity: AI can be used for threat detection, anomaly detection, malware analysis, and network security.

12.?? Human resources: AI can help with candidate screening, employee engagement, performance evaluation, and workforce planning.

13.?? Legal: AI can assist in legal research, contract analysis, case prediction, and document review.

14.?? Real estate: AI can be used for property valuation, market analysis, customer segmentation, and personalized recommendations.

15.?? Entertainment: AI can help with content creation, recommendation systems, audience segmentation, and trend prediction.

16.?? Sports: AI can assist in performance analysis, injury prevention, talent scouting, and fan engagement.

17.?? Environmental monitoring: AI can be used for air and water quality monitoring, wildlife tracking, climate modeling, and disaster response.

18.?? Smart cities: AI can help with traffic management, waste management, energy efficiency, public safety, and urban planning.

19.?? Social media: AI can assist in content moderation, sentiment analysis, influencer marketing, and personalized recommendations.

20.?? Personal productivity: AI can be used for time management, task automation, personal finance management, and goal setting.

And to think my prompt was in “natural language.” ?This list paints an amazing future.

But, there is potentially a scary future as well. So, again, using ChatBox AI (GPT-3.5) on my mobile phone:

Me:

What is scary about AI?

ChatBox:

There are several reasons why some people find AI technology to be scary or unsettling:

1.?????? Lack of control: As AI systems become more advanced, there is a concern that humans may lose control over these systems. This lack of control could potentially lead to unintended consequences or even dangerous outcomes.

2.?????? Job displacement: Many fear that AI technology will lead to widespread job loss as machines become more efficient at performing tasks traditionally done by humans. This could have significant economic and social implications.

3.?????? Privacy concerns: AI systems often rely on collecting and analyzing vast amounts of data about individuals. This raises concerns about privacy and the potential for misuse of personal information.

4.?????? Bias and discrimination: AI systems are only as good as the data they are trained on, which can lead to biased or discriminatory outcomes. This is particularly concerning in areas such as hiring, lending, and criminal justice where decisions can have serious consequences.

5.?????? Existential risks: Some experts warn about the potential existential risks associated with AI, including the possibility of super intelligent machines surpassing human intelligence and posing a threat to humanity.

Overall, the fear of AI stems from the unknown and the potential for these technologies to have far-reaching and unpredictable impacts on society.

When I think about the evolution in technology over recent decades, the unknown hasn’t stopped us, we (the collective societal we) have overcome and solved for the challenges that accompany all the innovations we now accept everyday as conveniences. I’ll pick one – the microwave. Today they come built-in to the homes built or remodeled new, and I remember the day my mom would not go near the microwave while it was heating leftovers for fear of those micro waves leaking out.

A lot of very smart people are working to address our AI anxieties at this very moment. Twenty years from now – an everyday convenience. Go forward with caution but go forward. ??