Artificial Intelligence has been around for many decades. In fact, it first emerged in 1956. However, AI has recently gone mainstream and become a buzzword, thanks in part to ChatGPT, a chatbot that mimics human speech. It was launched on November 30, 2022, and within just 5 days of its launch, it achieved 1 million users. By January 2023, it had amassed 100 million active users. Microsoft is one of its early investors and has been integrating AI Copilots into its line of products and Bing AI.
During a recent visit to my mother and sister, they expressed their concerns about AI potentially ruling the world. If you’ve watched the recent Mission Impossible: Dead Reckoning movie, the classic Terminator films, or read about ChatGPT’s ability to pass the bar exam in the US, you might share similar fears. This raises the question: Is the concept of AI eventually ruling the world farfetched, or is it a legitimate concern?
But really, how does AI work?
Before we answer this question, let’s explain how AI works in simple terms.
- Data Input: Developers begin by feeding a substantial amount of data into the AI system. This data can range from traffic images for self-driving cars and healthcare data for predicting patient outcomes to financial data for forecasting market trends. In the case of ChatGPT, it utilizes publicly available internet data to generate human-like responses.
- Learning: Developers program the AI to understand what to look for based on the data provided. They use algorithms to identify the best patterns within the input data.
- Decision-making: The AI then uses the identified patterns as a model to make decisions or predictions about future data.
- Feedback and Improvement: The AI can improve with feedback, but it relies on developers to guide its learning process, distinguishing between right and wrong to enhance its accuracy.
AI still has technical limitations
Considering how AI works, it’s evident that the notion of AI ruling the world, akin to Skynet in Terminator or the Entity in Mission Impossible, is implausible. AI operates based on the data and algorithms it’s given and is confined to specific tasks it’s designed for. AI lacks the ability to spontaneously learn or comprehend matters beyond its designated scope without further training or reprogramming. In essence, AI lacks the self-initiative to explore new areas of knowledge or make decisions outside its programmed boundaries. Unlike humans, who can change careers and acquire new skills based on personal choices, AI remains static without human intervention. Furthermore, achieving global domination would require processing an immense amount of diverse data and factors, making it exceedingly complex to train AI for such a purpose.
The good stuff
AI has the potential to revolutionize the way we live for the better. It can enhance road safety by perfecting algorithms to prevent accidents and improve health treatments and patient outcomes with AI-assisted medical professionals. AI has also been effective in preventing financial fraud and ensuring manufacturing quality, among other applications. Additionally, AI plays a pivotal role in aiding scientists in their quest to discover new planets. In real-life scenarios involving the identification of relationships and patterns within vast datasets and managing extensive repetitive tasks, AI simplifies human work.
The bad stuff
Despite its numerous advantages, AI usage also presents concerns and potential negative consequences. AI can be employed to propagate misinformation through deepfake audios and videos designed to impersonate individuals. Cybercriminals can utilize AI to conduct sophisticated cyberattacks, and AI-driven social media accounts and bots can disseminate propaganda and fake news. AI is susceptible to inheriting biases from the data it’s trained on, leading to discriminatory outcomes in fields like hiring, lending, and law enforcement. Consequently, implementing human oversight, ethical guidelines, and safety measures is essential to mitigate these risks.
How Microsoft addresses AI
Microsoft’s approach to AI serves as an example of responsible usage. The company removed emotion recognition from its facial recognition technology, recognizing its potential for manipulation and discrimination. Microsoft has stringent protocols for granting access to sensitive AI fields, requiring individuals to complete an application, and subjecting it to review by a dedicated team.
It won’t take over the world
AI is here to stay and will undoubtedly transform the way we work and live in the coming decades. It will influence our lives and industries in both positive and negative ways. For now, the concept of a General AI capable of performing diverse tasks and surpassing its designated scope to rule the world remains a distant possibility. Nevertheless, AI continues to evolve, and predicting its future capabilities is challenging. It is essential that companies offering AI technologies, like Microsoft, prioritize oversight, ethical guidelines, and safety measures to prevent the emergence of negative outcomes.