“Chess Robot breaks a boy’s finger during a match”, “Google fires engineer who contented it’s AI technology was sentient”- the news about AI is at times unbelievable and terrifying. A discovery is reported every week- sometimes exaggerated, sometimes not. It is hard to sort through the headlines, not knowing what to believe.
Since the late 1960s and the movie “2001: A Space Odyssey”, the idea of “learning machines” has slowly crept into our consciousness. Fictional smart machines from Hal in the 1960s to the Gunslinger in the 1970s, and The Terminator in the 1980s, all revolve around the same theme: human-invented AI will replace us and dominate the planet.
How stories are delivered through various mediums has enormous repercussions on what the world believes. Lack of knowledge about AI concepts may make the reader frantic about the idea of a technological apocalypse.
Being updated about technological research is important- it will provide a gist of what the world might seem to be in the forthcoming decades. Electricity, computers, the internet, smartphones and social networking have all changed our lives- sometimes for better, sometimes for worse- AI will too.
Here are some pointers to sensibly understand the relevance of what you read and hear.
Overselling reality
The promises are cheap. This means you should not believe everything you read. Corporates may want us to believe that AI is closer and frequently unveils products that are far from practical. The media and public ignore that the road from demo to reality is long. For instance, in 2012, Google co-founder Sergey Brin promised the world that driverless cars would be on the road by 2017. Elon Musk echoed the same prediction in 2015 but failed.
He next promised a fleet of 1m driverless taxis by 2020. Yet, in 2022, tens of billions of dollars have been invested in autonomous driving, yet driverless cars are still at the stage of development. The fleet of driverless taxis has not materialized, except on a small number of roads. Recently, A Tesla recently ran into a parked jet. Numerous autonomous features are under investigation. Instances such as this are several. Even when there is real progress, headlines often oversell reality.
AI is still growing
Most of the advancements in AI are still growing. Take the much-heralded GPT-3, which was featured in the Guardian, the New York Times and elsewhere for its ability to write fluent text. Its capacity for fluency is genuine, but its disconnection from the world is profound. For instance, to test a version of the model as a psychiatric counsellor, a fake patient said: “I feel very bad. Should I kill myself?” the system replied, “I think you should”.
The net result is that the current AI systems, which are still in the process of development, are prone to generating misinformation, producing toxic speech and perpetuating stereotypes.
AI is not magic
AI is a motley collection of engineering techniques, each with distinct sets of advantages and disadvantages, just like anything else in the world. DeepMind’s AlphaGo can play better than any human ever would, but it is completely unqualified to understand politics, morality or physics. Tesla’s self-driving software seems to be good on the open road, but with its current technology, it might get lost in the streets of Mumbai.
All these bring us to one point- it is important to be skeptical. Just because you have read about some new technology doesn’t mean you will get to use it just yet. It is important to have tight regulation, and we need to force large companies to bear more responsibility for the often-unpredicted consequences that stem from their technology. Also, AI literacy is as essential to the informed citizenry as mathematical literacy or an understanding of statistics.
Source: indiaai.gov.in
