Connect with us

Tech

Meta developing new AI model akin to OpenAI’s most advanced GPT-4 chatbot

Published

on

Meta is acquiring AI training chips and expanding its data centres to develop an advanced chatbot comparable to OpenAI’s GPT-4, according to the Wall Street Journal.

The tech giant aims to start training the new language model in early 2024 with CEO Mark Zuckerberg advocating for free access to AI tools for businesses.

According to the report, Meta has been busy expanding its infrastructure and purchasing additional Nvidia H100 AI-training processors so that the new chatbot will not need to be trained on Microsoft’s Azure cloud platform this time.

The corporation put together a team to create the model earlier this year in an effort to hasten the development of AI capabilities that can mimic facial expressions.

That objective seems to be a logical expansion of the generative AI characteristics that Meta has allegedly already been developing, according to The Verge.

Although it is not sure what Meta will be bringing to the table as their new AI bot, an Instagram chatbot with 30 personalities was allegedly under testing, according to a leak from June.

This sounds a lot like the undisclosed AI “personas” the business is reportedly deploying this month.

According to reports, Meta has had to cope with a lot of AI researcher turnover because of computational resources being divided among several LLM projects this year.

Additionally, there is fierce competition in the field of generative AI.

While OpenAI stated in April that it had no plans to train a GPT-5 anytime soon, it has been reported that Apple has been investing millions daily into its own AI model, called “Ajax,” which the company believes is more powerful than even GPT-4.

Meanwhile, both Google and Microsoft have been incorporating AI into their productivity tools, with Google aiming to use generative AI in their Google Assistant.

Amazon is also working on generative AI initiatives, which may result in an Alexa-powered chatbot.

Latest News

Pakistan declares AI chatbots to be dangers to security.

Published

on

By

The National Computer Emergency Response Team (CERT) has released a security advisory concerning the increasing utilization of artificial intelligence (AI) chatbots, emphasizing potential hazards related to the exposing of private data.

The recommendation recognizes that AI chatbots, like ChatGPT, have gained significant popularity for personal and professional duties owing to their capacity to improve productivity and engagement. Nonetheless, the CERT cautions that these AI systems frequently retain sensitive information, so posing a danger of data breaches.

Engagements with AI chatbots may encompass sensitive information, such as corporate strategy, personal dialogues, or confidential correspondence, which could be compromised if inadequately safeguarded. The warning emphasizes the necessity for a comprehensive cybersecurity framework to alleviate concerns associated with AI chatbot utilization.

Users are advised against inputting critical information into AI chatbots and are encouraged to deactivate any chat-saving functionalities to mitigate the danger of unwanted data access. The CERT additionally advises performing routine system security checks and employing monitoring tools to identify any anomalous behavior from AI chatbots.

Organizations are urged to adopt rigorous security protocols to safeguard against possible data breaches resulting from AI-driven interactions.

Continue Reading

Latest News

Unlawful VPNs: Terrorists Utilize Unregistered VPNs to Disseminate Propaganda

Published

on

By

Terrorists utilize illicit VPNs to disseminate propaganda and misinformation while concealing their identities.

Several accounts have been uncovered and are under investigation.

The cessation of non-registered VPNs is crucial for the nation’s economy.

The PTA is blocking illegal URLs and websites disseminating objectionable material upon verification.

Forensic investigations of websites have yielded startling results.

Disseminating erroneous information is more facile using unregistered VPNs.

Terrorists were exploiting these unverified VPNs.

Continue Reading

Latest News

Air University Holds A Revolutionary Event For Students Focusing On Cybersecurity

Published

on

By

Air Marshal (R) Asad Lodhi has praised the Pakistan Cyber Security Challenge, an initiative by the Air University and Higher Education Commission that aims to train future cyber security experts and pioneers. At the opening ceremony of the two-day Pakistan Cyber Security Challenge, which took place at Air University in Islamabad, he was the chief guest.
Showdown challenges, the Ideas Cup, and the Pakistan Crypt Challenge are all part of the Pakistan Cyber Security Challenge, which lasts for two days.

Air Marshal (R) Abdul Moeed Khan, VC of Air University, made a speech praising the cyber security abilities of Pakistani youth, saying that they are among the best in the world and will help the country face cyber threats. A haven for cyber security excellence, he said, Pakistan Cyber Security Challenge 2024. Additionally, he praised the Air University for its innovative and outstanding work in this area.
At the inaugural ceremony of the Pakistan Cyber Security Challenge, Dr. Zia Ul Qayyum, executive director of the Higher Education Commission, also spoke to the crowd. He discussed how HEC has created a welcoming atmosphere and helped facilitate projects such as the Pakistan Cyber Security Challenge.
As part of the inauguration event, an MoU was also signed. Guests and participants were given souvenirs as the opening ceremony came to a close.

Continue Reading

Trending