Technology giant Google is going to end beloved feature of Gmail.

Gmail users who use the website’s HTML Basic View feature will no longer have this feature available from January 2024.

According to a notice in the company’s support patch, Gmail will only be available in Standard View from that date.
HTML BasicView is originally a 2007 design by Gmail that users often use over slow Wi-Fi. Standard view features such as chat box, spell check or contact importing are not loaded on this mode.

However, Pratik Patel, an expert in blind technology, warns that the feature being retired is a major setback for blind and partially sighted people.

He said that many blind people who use this mode of Gmail will not only be confused but also unhappy.

HTMLView makes the screen much easier for blind people to use. The standard interface is now available for use by the blind, but many people are left with the basic HTML version because that’s what they know how to use.

Pratik Patel added that it is not that standard interfaces are not available. But some people will not be ready to use this mode.

Imagine a cutting-edge AI system that can teach robots to do things they were never specifically trained for. Well, that’s precisely what Google DeepMind has achieved with their new creation called RT-2. This groundbreaking model learns from both web-based data and information gathered from real-world robotics experiences, enabling it to give simple yet effective instructions to robots.

Testing Phase of RT-2:

In the testing phase, RT-2 was challenged with tasks that were completely new to it, not part of its initial training. For example, it was asked to arrange oranges in a matching bowl, something it had never done before. To tackle these unfamiliar tasks, RT-2 tapped into the vast knowledge it had accumulated from the web data it learned. And impressively, it succeeded in completing these actions with a 62% success rate – that’s twice as good as its predecessor, RT-1.

Ability to work on New Commands:

But what makes RT-2 truly remarkable is its ability to adapt and think on its feet. It can figure out new commands and respond effectively, even when those tasks were never part of its training data. For instance, when asked to find an object that could serve as a hammer, RT-2 astutely identified a rock as the most suitable option.

Moreover, RT-2’s cognitive prowess goes beyond single-step problem-solving. It can reason through multiple stages, devising clever solutions to complex challenges. In one test, it ingeniously concluded that a rock would make a perfect makeshift hammer, displaying a level of reasoning similar to how humans would approach the problem. Another time, when commanded to push a bottle of ketchup towards a blue cube, RT-2 achieved the objective, despite having no prior knowledge of any objects except the cube.

This breakthrough by Google DeepMind signifies a major leap forward in the realms of artificial intelligence and robotics. RT-2’s capability to learn from web data and apply that knowledge to real-life tasks opens up exciting possibilities for the future of robotics. As the technology continues to evolve, we could witness the emergence of versatile and helpful robots that can seamlessly integrate into various human-centered environments. While there’s still much work to be done, RT-2 offers a tantalizing glimpse of what’s to come in the fascinating world of robotics.

Some Questions You may want to Know About RT-2:

1. What is RT-2?

RT-2 is an advanced AI model developed by Google DeepMind for robotics. It can teach robots to perform tasks beyond its initial training data, thanks to its ability to learn from web data and apply that knowledge to real-world scenarios.

2. How does RT-2 learn to perform new tasks?

RT-2 learns from a combination of web-based data and information collected from real-world robotics experiences. It uses this diverse knowledge to give robots simple instructions for performing various tasks.

3. What is the success rate of RT-2 in performing new tasks?

In tests, RT-2 achieved a 62% success rate in carrying out tasks that were not part of its initial robotic training data. This is twice as successful as its predecessor, RT-1.

4. Can RT-2 understand and communicate with robots?

Yes, RT-2 can effectively communicate with robots by providing them with clear and actionable instructions. It can direct robots to carry out tasks that it was not explicitly trained for.

5. Does RT-2 demonstrate reasoning capabilities?

Absolutely! RT-2 showcases remarkable reasoning capabilities, allowing it to figure out new user commands and tasks that were not present in its initial training data. It can even perform multi-stage semantic reasoning to solve complex challenges.

6. How does RT-2 adapt to new situations?

RT-2’s adaptability comes from its ability to draw upon the diverse knowledge it acquired from web-based data. It can translate this knowledge into useful instructions, enabling it to tackle new and unfamiliar tasks effectively.

7. What is DeepMind’s view on RT-2’s impact on robotics?

Google DeepMind sees RT-2 as a significant breakthrough in artificial intelligence and robotics. They believe it brings us closer to a future of helpful and more general-purpose robots that can operate in various human-centered environments.

8. How can RT-2 enhance interactions with robots?

RT-2’s ability to teach robots new tasks and communicate effectively with them can significantly enhance human-robot interactions. It opens up possibilities for robots to be more adaptable and helpful in diverse situations.

9. What future possibilities does RT-2 present for robotics?

RT-2’s success in learning from web data and applying that knowledge to real-world tasks offers exciting possibilities for the future of robotics. As the technology advances, we could see robots becoming more versatile, capable, and seamlessly integrated into our daily lives. The potential for helpful robots in various human-centered environments is just within reach.

Elon Musk’s announcement to fire Twitter’s “Bird”. Replaced By Twitter X.
Elon Musk, the owner of Twitter, has announced that the well-known logo of the blue bird (sparrow) in the social media platform is now going to be a thing of the past, Elon Musk will say goodbye to the bird and other birds soon.

According to details, Twitter’s iconic bird logo is going to change, Elon Musk confirmed that Twitter’s logo will soon become an “X”. This will be the biggest change of Twitter since Elon Musk’s ownership in October 2022.

Apart from this, he also sounded the alarm for the old employees. The birds are referring to former Twitter employees who founded the company with the Twitter bird logo.

Recall that in April 2023, announcing the appointment of the new chief executive of Twitter, Landa Yacarino, Elon Musk said that she will work to transform the platform into “X”, which will be a super app.

Earlier in April this year, Twitter’s traditional logo was temporarily replaced with a Doggy Coin dog logo.