For the Love of Art

Microsoft is putting ChatGPT through gruelling testing. While others are busy toying with the chatbot for generic purposes, the big tech is using ChatGPT to build some formidable products like Visual ChatGPT. The latest competitor to DALL.E 2, Midjourney and Stable Diffusion, Visual ChatGPT allows users to command the platform in the natural language to generate relevant pictures.?

Microsoft used ChatGPT and Stable Diffusion, an open-source text-to-image model, to create a platform that bridges the gap between text-to-image and natural language generation. The need for Visual ChatGPT emerged as the ‘prompt’ technique was turning out to be a complicated method to achieve desired results on popular text-to-image tools. It required a comprehensive knowledge base of what to prompt for.?

Visual ChatGPT functionality is facilitated through a system architecture that utilises a ‘prompt manager’ to share information between various visual foundation models, such as Stable Diffusion, ControlNet, BLIP, and ChatGPT itself.

With tools like Visual ChatGPT, text-to-image models may become interoperable and compatible with other AI tools. These technologies can assist in bridging the divide that previously existed between language models and picture-generating models. We might be able to further improve the capabilities of these sophisticated models by applying technologies like prompt management.

Read the full story?here.?

Exchange of Business Models

It’s a swap. OTT platforms are going the social media way and social media is going the OTT way exploring each other’s business models. While OTT players integrate ads into their platforms, social media has taken to subscriptions. The biggest issue OTT faced was the non-paid users, who consumed the content, but were not ready to pay for it. On the other hand, social media platforms had users who readily paid for a personalised experience.

Read the full story?here.

No Innovation for Indian Trucks

Tesla, Daimler, and Aurora are investing heavily in self-driving trucks. In the past few years, some major breakthroughs have also been achieved in the trucking industry. But the development is only limited to international markets. There’s little for the Indian trucking industry to cheer about. So far, the industry has witnessed small innovations in the form of aggregation of trucks and in-app booking facilities.

Read the full story?here.

AGI is No Good

Professor Michael Irwin Jordan from UC Berkeley believes that creating AGI (artificial general intelligence) is a "lazy person's aspiration”. Instead, he suggests focusing on developing AI systems that can bring music and choice into peop's homes, arguing that AGI is not necessary or sufficient for achieving such tasks.

Read the full interview here.