High 5 tales of the week: Nvidia and OpenAI updates, 2023 predictions for cybersecurity, and extra

As we enter December, marking the 12 months’s finish and start to sit up for what’s on the horizon for the tech business as we method 2023, consultants throughout a variety of firms have unveiled their predictions, insights and even a look at up-and-coming improvements.

Even within the midst of a doable recession, it’s clear that synthetic intelligence (AI) consultants anticipate innovation to proceed — particularly in relation to generative AI. Actually, this week, Jensen Huang, CEO of Nvidia shared insights on the intensive use instances generative AI might supply for the metaverse, together with the rising demand to fill the digital world with 3D and graphic belongings. 

Relatedly, OpenAI, the corporate behind generative AI instruments DALL-E and GPT-3, famous this week that it’s engaged on a brand new AI-powered language mannequin which is able to develop into a part of the GPT-3 household. Reportedly the mannequin, GPT-3.5, will have the ability to deal with extra complicated directions and in flip produce greater high quality outcomes. 

AI advances within the medical area additionally made the information this week. Nvidia introduced it’s working to raised combine AI fashions into medical workflows and likewise aiming to deliver simplicity to medical imaging  — an space that Harvard Medical College’s AI analysis workforce can be engaged on individually. 

With the numerous technological advances throughout industries comes discussions about cybersecurity vulnerabilities as nicely. 2022 was a 12 months of many wins and likewise issues for IT and safety professionals. Simply as a Twitter API breach uncovered the knowledge of 5.4 million customers, which was just lately made accessible totally free by hackers — safety continues to be an ever-changing panorama with quickly shifting targets. To achieve perception into the largest threats, VentureBeat spoke to 31 CISOs t from firms together with Google, IBM, Microsoft, AWS and dozens of others about what they’re paying shut consideration to heading into the brand new 12 months.

Right here’s extra from our prime 5 tech tales of the week:

1. How generative AI might create belongings for the metaverse
   Jensen Huang, CEO of AI and graphics chipmaker Nvidia, believes that generative AI will likely be transformational and it’s simply getting began. Certainly one of its largest purposes might be with the metaverse, which has large calls for for content material as builders have to fill out digital worlds with 3D belongings. Corporations like Steady Diffusion, Promethean AI and Ludo AI are utilizing these applied sciences to routinely generate paintings and different belongings for gaming and metaverse purposes.

   Many metaverse firms are hoping that generative AI will assist present the assets to assist them construct out their worlds. Huang believes you will notice progress whenever you enter an increasing number of prompts — comparable to textual content to flesh out an idea — and the idea imagery will get higher and higher.

2. OpenAI debuts ChatGPT and GPT-3.5 sequence as GPT-4 rumors fly 
   As GPT-4 rumors fly round NeurIPS 2022 this week in New Orleans (together with whispers that particulars about GPT-4 will likely be revealed there), OpenAI has managed to make loads of information within the meantime.

   On Monday, the corporate introduced a brand new mannequin within the GPT-3 household of AI-powered giant language fashions (LLMs), text-davinci-003, a part of what it calls the “GPT-3.5 sequence,” that reportedly improves on its predecessors by dealing with extra complicated directions and producing higher-quality, longer-form content material.

3. Twitter API safety breach exposes 5.4 million customers’ information
   In July, this 12 months, cybercriminals started promoting the consumer information of greater than 5.4 million Twitter customers on a hacking discussion board after exploiting an API vulnerability disclosed in December 2021. 

   Not too long ago, a hacker launched this info totally free, simply as different researchers reported a breach affecting thousands and thousands of accounts throughout the EU and U.S.

4. The way forward for AI and medical imaging, from Nvidia to Harvard
   Nvidia introduced this week on the annual assembly of the Radiology Society of North America (RSNA) that MONAI, an open-source medical-imaging AI framework accelerated by Nvidia, is making it simpler to combine AI fashions into medical workflows with MONAI Utility Packages (MAPs), delivered by means of MONAI Deploy.

   Nvidia and King’s School London launched MONAI in April 2020 to simplify AI medical imaging workflows. This helps remodel uncooked imaging information into interactive digital twins to enhance evaluation or diagnostics, or information surgical devices. The event and adoption of the platform now has over 600,000 downloads, half of those within the final six months.

5. 31 CISOs share their safety priorities and predictions for 2023
   2022 was a pivotal 12 months within the cyberthreat panorama. Securing the software program provide chain and the open-source software program ecosystem, implementing zero belief, and educating workers in regards to the dangers of social engineering and phishing makes an attempt are simply a number of the areas that CISOs are evaluating to mitigate potential dangers.

   Executives from Google, Microsoft, AWS, IBM and extra advised VentureBeat what they predict for the cybersecurity sector within the coming 12 months and what’s prime of thoughts for them going ahead.