What are 10 problems for society that will be caused by AI in the next 5 years?

Scammers will begin to make spam attacks more sophisticated and will co-opt GPT's creativity for nefarious scams, creating convincing advertisements that slip by spam filters or launching more sophisticated phishing attacks. They may also merge the VoiceCloning and ChatGPT systems to create interactive scamming systems over the phone, making it easier to take advantage of vulnerable people.

With the rise of AI, emulating competence and passing oneself off as an expert is becoming easier. For instance, on platforms like Reddit and Quora, more and more responses are generated by AI systems like ChatGPT. On websites like Fiverr, anyone can use generative AI to offer fake quality services, slap on a few certifications, and claim to be an expert, just like selling tap water and calling it Evian.

Moreover, people have been caught cheating during interviews and coding tests by using ChatGPT to generate answers. As AI systems become more advanced, these schemes may become more sophisticated. For example, imagine using Google Glasses paired with voice-to-text technology to take user questions and convert them to summarized answers.

In the worst-case scenario, con artists could run elaborate schemes, such as Ponzi or crypto scams, using AI-generated content to deceive their victims.

The widespread adoption of AI may lead to large-scale disruption of knowledge workers across various industries. This could result in job losses, lower wages, and increased inequality, particularly in countries where social safety nets are weak.

The use of GPT technology by corporations to find and create loopholes in bills being passed in Congress can lead to laws that benefit them at the expense of the public. This could exacerbate inequality and erode trust in the democratic process.

Politicians using GPT technology to look for loopholes and massage language to pass bills under the radar may result in laws that are not in the best interest of the public. This could undermine public trust in government and lead to social unrest.

Law offices may leverage GPT to become more creative in defending clients or using clauses in favor of clients for good or bad. They may also look for loopholes in laws to improve offshoring operations.

Drug cartels could use AI to manipulate public opinion and automate and streamline various criminal activities, including planning and executing drug trafficking, money laundering, and other illicit operations.

Nation-states using AI to launch sophisticated cyber-attacks can cause significant damage to critical infrastructure, disrupt essential services, and compromise sensitive information.

LLM companies with lax security may be vulnerable to data leaks that could cause serious privacy concerns. There have already been cases where some users were able to see the prompts and responses of other users. Hackers may also use private information to blackmail people for money or even hold the LLM company

Providers of large language models may use the data generated by these systems to gain a competitive advantage over others in the market. This could lead to the concentration of power in the hands of a few companies, limiting innovation and creating barriers to entry for new players.