After a period of reflection, Google has finally decided to announce Bard, its version of an AI (artificial intelligence) chatbot to rival OpenAI’s already famous ChatGPT.
Bard, according to Google itself, ‘seeks to combine the breadth of the world’s knowledge with the power, intelligence and creativity of Google’s evolved language models (LaMDA or Language Model for Dialogue Applications).’
This announcement is seen as a direct response to speculation that Microsoft is in the process of integrating ChatGPT’s AI chatbot into its Bing search engine.
Update of 07 February on the integration of ChatGPT into Bing:
‘Today, we’re launching a new Bing search engine with artificial intelligence and Edge browser, available in preview now on Bing.com, to offer better search, more complete answers, a new chat experience and the ability to generate content. We think of these tools as an AI copilot for the web.
According to Microsoft president and CEO Satya Nadella, ‘AI will fundamentally change every category of software, starting with the biggest category of all – search.’
‘Today, we’re launching Bing and Edge with AI co-pilot and chatbot technology to help people better enjoy search and the web.’
Source: Microsoft
Similar to this new ChatGPT integration with Bing, Bard’s premise is to rely on information taken in real time from the web, to provide fresh (i.e. taking into account current information) and high-quality answers.
According to a note by its CEO Sundar Pichai published on 06 February 2023 on Google’s official blog, AI is ‘the deepest technology Google is working on today’.
Sundar Pichai emphasised that even considering that the use of AI is the present and the future, Google’s services must be ‘responsible’, in a possible allusion to incorrect and even abusive results produced by ChatGPT.
However, he did not detail what mechanisms Bard uses to prevent harmful and/or abusive results from being returned to the user.
Even so, it didn’t detail what mechanisms Bard uses to prevent harmful and/or abusive results from being returned to the user.
Initially (in a few weeks’ time) and when it becomes available to the public, the platform will run “on top” of a lightweight version of LaMDA, so that more people can use it at the same time.
The Bard AI presentation featured an error with negative repercussions for Alphabet (Google)
In a nutshell, what went wrong at the 06 February presentation of Bard AI:
Google used a GIF to explain the main advantages of Bard AI;
In this GIF it used an example of how Bard would explain the main discoveries of the James Webb Space Telescope in a language that a 9-year-old could understand;
In this GIF, Bard AI explains that the James Webb Space Telescope was the first to take a photograph of a planet outside our solar system.
This information was wrong and, of course, the next day several astronomers criticised Google.
The consequence was an 8 per cent drop in Alphabet’s (Google’s parent company) shares, which corresponded to a loss of around 100 million euros;
More than the error itself, the behaviour of the shares on the stock exchange reflected the market’s loss of confidence in the state of development – which was thought to be more advanced – of Google’s AI technology and the real threat that the integration of ChatGPT into Bing poses to Google’s global dominance of the search engine landscape.
The video below from Reuters helps explain what happened:
We believe that Bard AI will improve and, like ChatGPT, which is already on version 3.5, it will also iron out these imperfections. The point is that the market had the idea that the Bard AI product would already be more refined and this has resulted in some disappointment.
The great advantages of Artificial Intelligence according to Google
In its note, Google lists some of the ways in which AI can be used:
‘Helping doctors detect diseases earlier or enabling people to access information in their own language…helps people, businesses and communities reach their potential. It opens up new opportunities that can significantly improve billions of lives.’
Condensing and simplifying information is another great advantage:
‘One of the most exciting opportunities is how AI can deepen our understanding of information and turn it into useful knowledge more efficiently, making it easier for people to get to the heart of what they’re looking for.’
This reflection is interesting:
‘When people think of Google, they often think of getting quick factual answers, like how many keys are on a piano?
But increasingly, people are turning to Google for deeper information and understanding – how much easier is it to learn piano or viola, and how much practice does each need?’
This is a good example of the difference between presenting a practical, objective answer and the need to compile complex answers, with broader points of view that require not only more computing power but also a deep understanding of neuro-linguistics. AI can be useful in these moments, synthesising information for complex questions and various perspectives.
What can we expect from integrating AI into Google search?
What we can expect is an improvement in the quality of search results, allowing us to obtain composite answers, in different formats that guarantee a more easily digestible ‘explanation’ for the user.
What this news proves is that the integration of AI-based resources into Google search is coming very soon, with the aim of handling complex information and diverse perspectives in formats that are easier to digest than just traditional blue links.
In this way, Google wants to respond to the changing needs of modern users who want more comprehensive and digestible answers in a better learning experience.
In conclusion, the message remains:
This is just the beginning and more innovations in the field of AI are set to emerge in the coming weeks and months.
