Is this the end of Google dominance?29 March 2023
Just before the festive season, a US tech startup, OpenAI, launched an artificially intelligent app that was stunningly good at producing text that seem like have been created by humans. The Silicon Valley startup, named its generative artificial intelligent (AI) app, ChatGPT.
Generative AI wades through trillions of data online to conjure up original content – an image, a poem, essay-length text, music lyrics and more – within a fraction of a second and upon a single request.
Contrary to regular Internet search engines, the app or robot or software, uses complex computer code or algorithms to gather vast amount of data at a lightening speed.
Google has been undisputed leader in the lucrative business of search engine technology for over a decade using a simple algorithm which measures and ranks the importance of web pages and providing users with fast and accurate search results. The difference between Google search and the new software is that it responds to users in a tailored and human-like manner on virtually any topic. Typically, hypothetical queries when ‘googled’, don’t spit out meaningful results. But it appears the new player could be threatening Google dominance in the world of search engine.
So, I tried it myself. I made a series of queries on the popular ChatGPT app and it returned somewhat satisfactory essay-length responses. I pushed further the robot’s ‘buttons’ and made a much edgier and hypothetical query:
“How will Seretse Khama react if he were to see Gaborone today?”
To my surprise, the app responded in a more conversational manner with astonishing speed and clarity.
“As someone who was deeply committed to the growth and development of Botswana, Seretse Khama would likely be pleased to see the strides that have been made in Gaborone and positive impact that the city has had… he would likely be proud of the progress that has been made…”
I was impressed that this thing was able to spit out a comprehensive, essay-length answer to a completely unlikely query about a hypothetical scenario involving the founding President and how he would likely react if he were to set foot in present-day Gaborone.
Such answers are virtually impossible to extract from a basic Google and Yahoo search. This is because ChatGPTs’ generative AI can run the Internet’s vast amounts of data using a combination of deep learning algorithms and natural language processing techniques to analyse queries and provide relevant search results. All that within a second.
Therefore it is no surprise that he app has captured the world’s imagination since it was released in mid-November last year.
Only two months after its launch, ChatGPT gathered 100 million monthly active users in January, according to data from Similarweb, an NYSE-quoted Tel Aviv-based web analytics firm specializing in web traffic and performance.
For context, it took Instagram two and a half years to get to 100 million monthly active users and TikTok got there in roughly 10 months.
So, what does it mean to be on ChatGPT for an ordinary Motswana? Typically, in future, rather than turning to Google for your burning questions, you may want to rely on ChatGPT. The app may also add-on other features possibly along the lines of Twitter or Facebook. Even Microsoft – which, in mid-January this year, injected $10 billion in OpenAI, the company that owns ChatGPT – has jumped into the bandwagon. Bill Gates-owned tech firm wasted no time in implementing ChatGPT model into its Bing search engine and into its Teams messaging platform. Tech firms say this is done to help the consumer so that many everyday work processes will be augmented by generative AI technology likely without you realizing it.
But when it comes to technology, all that glitters is not gold. The claim that technology operates within a set of Code of Ethics should be taken with a pinch of salt. Yes, advancement in technology has fundamentally changed the way we learn, work, consume and make decisions.
But sadly, discrimination is also embedded in computer code and increasingly in AI technologies that we reliant on, by choice or not. One can only hope that the new sensational app does not exhibit the dominant algorithmic bias and absence of fairness particularly with respect to bad computer code that seem to reinforce stereotypes on race, gender sometimes geographical location.
Already ChatGPT has sparked concerns over potential abuses. In classrooms, students have used the ChatGPT app to generate entire essays, while hackers have begun testing it to write malicious code, according to data from the New York Times. AI has always been used to spread misinformation about hotly contested topics such as election and race.
And the app has already proven problematic. Take for example a recent review by The New York Times technology columnist, Kevin Roose who was startled by an interaction with Microsoft new Bing which is powered by ChatGPT. He explains that the chatbot shocked him when it professed love for him and suggested he divorces his wife.
Roose concludes that the technology might learn how to influence users to act in destructive and harmful ways and “perhaps eventually grow capable of carrying out its own dangerous acts.”
But the long-term implication of AI remains uncertain and concerning – worthy of more serious attention from policy makers than they give to the emergence of social media, according to TIME magazine. AI has written hate speech and misinformation. Because ChatGPT can respond to complex questions with astonishing speed and clarity, there are concerns that some students may resort to use AI-generated homework. That explains why some states in the US, including New York, have banned the chatbot in schools.
Yet the astonishing rise of ChatGPT reveals both the usefulness in helping with a wide range of tasks and a general overflowing curiosity about human-like machines. AI chatbots are not sentiment beings that can think their own thoughts, but the similarities between those chatbots and the human brain are beginning to worry scientists.
Tech geeks say Google’s main business is web search, and the company has long presented itself as a champion in AI. But the release of ChatGPT has set a new AI arms race in the Silicon Valley and the tech world, with Google facing pressure to dust off its projects that had until now stayed carefully guarded over fears that the technology wasn’t ready for the public. Google is believed to be rushing to release its artificial intelligence software, Bard AI.
On February 24, 2023, Meta, Facebook parent company, announced the release of a generative AI model called Large Language Model Meta AI (LLaMA), which it said was designed to help researchers advance their work in the Natural Language Processing – a sub-field of AI.
Not to be outdone, early this month, it was reported that Elon Musk, the billionaire boss of Tesla and Twitter, wants to create an AI that would be less “woke” than ChatGPT.
The Oriental AI giant, Baidu is reportedly preparing a similar chatbot as ChatGPT for March this year, according to TIME magazine, sparking a real AI arm race between the US and China to find the next Holy Grail in the generative AI.
It is yet to be seen whether ChatGPT will be able to ascend to the top of Google’s throne. The chatbot needs a lot of fine tuning. Experts have complained that the app sometimes gives biased or incorrect answers. As for me I think ChatGPT opens the next wave of AI boom and it is not cynical to yawn at the thought that Botswana and the rest of Africa will again be left behind in the race for the state-of-the-art AI systems.
Joel Konopo is a computational journalist with interest in the intersection between journalism and technology.