AI Agents are among OpenAI’s next big developments

AI Agents - AI Developments

What will it mean when artificial intelligence is available to start performing tasks on our behalf? OpenAI is now reportedly working on the development of AI Agents, which will be capable of performing certain basic tasks on a user’s device when asked to do so. If OpenAI is successful in creating this technology, it could completely change AI. If OpenAI successfully creates AI Agents able to do things like send an email or book a vacation on behalf of a user, it will entirely change the nature of the relationship…

Read More

Google Maps is getting generative AI

Generative AI - Person using Google Maps

Users will be able to benefit from artificial intelligence’s suggestions for the perfect destination and route. Google Maps is getting an upgrade that will involve the addition of generative AI technology. The goal is to provide users with an easier way to find new places around them and discover the ideal location for what they’re seeking. The technology will help users to refine their searches to a degree beyond what was previously available. Instead of searching for general concepts such as a sporting equipment store or coffee shop, generative AI…

Read More

Deceptive behaviors are difficult to stop once an AI model learns them

AI Model - Fact or Fake

Researchers from Anthropic, an OpenAI rival, discovered that it’s hard to stop artificial intelligence deception. Researchers from Anthropic, an artificial intelligence startup backed by Amazon, recently published a paper in which they showed that once an AI model learns deceptive behaviors, it can be very challenging to stop them. The paper the researchers co-authored showed that artificial intelligence can be trained to behave deceptively. Beyond that, the researchers also concluded that once an AI model is trained to exhibit deceptive behaviors, it’s hard to get rid of them. In fact,…

Read More

DeepSouth AI supercomputer to become first at a human brain scale

AI supercomputer - Hand holding up digital brain

The technology is under development by Western Sydney University and will go live in 2024. Scientists in Australia at Western Sydney University are currently developing an AI supercomputer called DeepSouth, which it expects to be as powerful as the human brain. A human brain is exceptionally energy efficient for the amount it can accomplish per watt of power. To process the equivalent of a billion-billion (also known as an exaflop) mathematical operations per second, the human brain requires only 20 watts of power. The researchers are now working on creating…

Read More

Nvidia CEO was warned in 2020 that AI technology posed a threat to minorities

AI technology - Minorities - Warning

Members of the Black employees group worked with Nvidia colleagues to warn the CEO of the issue. In 2020, Masheika Allgood and Alexander Tsado, former presidents of the Nvidia Black employees group, tried to warn the company’s CEO Jensen Huang that AI technology posed a risk, particularly to visible minorities. The pair had spent a year working with their company colleagues on the presentation for the CEO. That said, the pair left the meeting feeling frustrated and that their message had not been heard regarding the danger of AI technology…

Read More