Artificial intelligence (AI) has demonstrated its usefulness across various industries and will likely impact the future in ways we can’t even imagine. There are equal parts apprehension and eagerness to see what AI could do for us, but we can never be 100% sure until the time comes. For now, here are some predictions analysts have on the benefits and dangers of AI.

AI will create millions of jobs

“Robots are taking our jobs!” is the familiar battle cry of those against the growth of AI. However, the truth is that the increased implementation of AI is also creating demand for workers with AI or machine learning-related skills. Data show that 71% of companies are now seeking employees with experience in data annotation, computer programming, AI testing, etc. Many institutes also offer courses and specializations on these topics, which can help buff up job-seekers resumes.

There will be increased regulations on AI

There is much debate whether specific applications of AI constitute privacy breaches. For example, social media takes advantage of AI and users’ lax privacy settings to develop targeted ad schemes that they sell to corporations and businesses. This leads to user data unknowingly becoming products or bargaining chips for these companies. Additionally, there is concern about AI’s applications in government espionage and weapons manufacturing. The issue with technology developing faster than we can regulate is that cases like this do not have the necessary legislation to protect private individuals. We’ll likely see more laws passed involving AI and its use.

We will need backdoors and fail-safes when AI goes rogue

Often, people have too much faith in their creations. This arrogance has led to great tragedies, such as the Titanic. The engineers’ failure to put enough lifeboats on the ships resulted in numerous avoidable deaths, and history could repeat itself if we do not tread carefully. The doomsday scenario of AI becoming sentient and turning against us is highly unlikely, but what can happen is AI misinterpreting the input it’s given. An example of this is Microsoft’s AI chatbot, Tay. This AI was an experiment in how well AI could hold conversations on Twitter. However, some users began feeding the AI with hateful rhetoric, forcing Microsoft to shut it down. With how easy it is for AI to spin out of control, we need to ensure that we have protective measures available just in case.

AI will reduce disease-related fatalities

AI is already doing much for contemporary healthcare, but their potential will only grow as they become more innovative and efficient. Scientists are developing AI models used as early detection warnings for COVID-19, cancer, and other potentially fatal diseases. AI can rapidly diagnose thousands of patients, which can help them get treatment early, saving their lives.

Concerns about AI are valid. Even top AI researchers have reservations about the trajectory of AI and how easily things can go wrong. However, we shouldn’t let the fear overcome the truth that AI can improve human life.