Between the Devil and the Deep Blue Sea

Between the Devil and the Deep Blue Sea

Stuck like a mole in a snake’s mouth. This is loosely based on a famous Hindi proverb used to describe a difficult situation when an opportunity becomes a bane. Stack Overflow, the public Q&A platform for developers, is currently experiencing something similar where it’s unable to decide whether to adopt the new generative AI technology or let go. 

Stack Overflow first scoffed at the chatbot and then banned it from the discussion forum — any post generated by ChatGPT and posted on the discussion platform was taken down by the platform. The forum, however, realised users were quietly moving away. Reports from SimilarWeb authenticate the fall in the number of users at Stack Overflow after the release of ChatGPT in November. There was a 12% decrease in the number of website visits, which went from 279 million to 247 million in December. After just a slight rise in January to 249 million, the website visits dropped even lower in February, to 239 million. 

However, Stack Overflow casually blames the decline on the holiday season. 

Perturbed by the situation, Stack Overflow mellowed its stance and has decided to go the gen AI way. It will use a considerable corpus of user-generated content to train an LLM. The LLM-based platform will reportedly be trained on the knowledge base of over 50 million questions from Stack Overflow, along with proprietary data from the Stack Overflow for Teams service. 

While Stack Overflow was just about letting out a sigh of relief with its newly-minted LLM, the plan has backfired. Now, the community is upset! They feel that the move to adopt LLM for Stack Overflow is pointless. It’s a community-driven platform and adding generative AI to it will drive away the real contributors from it, they say.

This hero of the hour, generative AI, has put the platform in a tight spot. Stack Overflow adopted gen AI to placate the losing AI-enthu users, but, now, the main community is irked by the new decision. When the platform tries to pacify one party, the other gets angry. Phew! This does sound like a real plight.

Read the full story here.


AI Killed ETL

Organisations are moving away from ETL due to the emergence of AI agents like AutoGPT and AgentGPT, as well as a preference for unstructured data lakes over data warehouses. Traditional ETL pipelines have transformed to meet the demands of modern analytics use cases, but they can still be viewed as a hindrance to better performance and require substantial investments in hardware, software, and personnel resources. AI agents can automate data processing and analysis, providing faster and more accurate results.

Read the full story here.


Google Search Faces New Threat 

It looks like Google is heading towards its end, for the process has already begun. Google Search is not a default player in the market. To remain so, it invests insane wads of dollars. The search engine pays about $12 billion a year to Apple to be its default search engine on all devices. Similarly, it pays about $3 billion annually to Samsung for the favour. 

But, according to recent reports, Samsung is contemplating switching to Bing as its default search engine, with Apple likely to follow suit. If this happens, Google Search will definitely have to look for a better strategy to remain the Number 1 player. Else, Bing won’t take much time catching up.

Read the full story here.


Is Sundar Pichai Lying?

Alphabet CEO Sundar Pichai is on an interview spree. In all his interviews with the media, he is talking about his big dream with Google and how the company’s AI future is going to be. In one of the interviews, Pichai disclosed the capability of its chatbot Bard and how it started teaching itself skills that it wasn’t trained on. He explained that one Google AI programme adapted on its own upon being prompted in the Bengali language, which it was not trained to know. 

His statement, however, was contested by a former Google researcher who took to Twitter and said that Google’s PaLM, the AI model which is the forerunner to Bard, had, in fact, been trained to understand Bengali. 

Read the full story here.

To view or add a comment, sign in

Insights from the community

Others also viewed

Explore topics