A good open world game is filled with little details that add to a player's sense of immersion. One of the key elements is the presence of background chatter. Each piece of dialog you hear is known as a "bark" and must be individually written by the game's creators — a time consuming, detailed task. Ubisoft, maker of popular open world gaming series like Assassin's Creed and Watch Dogs, hopes to shorten this process with Ghostwriter, a machine learning tool that generates first drafts of barks. 

To use Ghostwriter, narrative writers input the character and type of interaction they are looking to create. The tool then produces variations, each with two slightly different options, for writers to review. As the writers make edits to the drafts, Ghostwriter updates, ideally producing more tailored options moving forward.

The idea here is to save game writers time to focus on the big stuff. "Ghostwriter was created hand-in-hand with narrative teams to help them complete a repetitive task more quickly and effectively, giving them more time and freedom to work on games' narrative, characters, and cutscenes," Ubisoft states in a video release.  

Ubisoft touts Ghostwriter as an "AI" tool — the big thing at the moment with seemingly every company, from Google to Microsoft, hopping onboard the AI train. 

Like similar tools, though, the question is how to get people — namely staff — to actually use it. According to Ben Swanson, the R&D scientist at Ubisoft who created Ghostwriter, the biggest challenge now is integrating the tool into production. To better facilitate this, the production team created Ernestine, a back-end tool that facilitates anyone to create new machine learning models in Ghostwriter.

If Ghostwriter proves effective, writers should be able to spend their time and energy building more detailed and engaging gaming worlds to explore. 

To have any hope of making its Horizon Worlds VR social network catch on, Meta has to give potential users a reason to go (and stay) there. Its latest attempt to do that is something called quests that lets users complete in-game missions to earn (virtual) swag like clothing, the company announced in an update spotted by The Verge. 

The feature (which doesn't appear related to the Quest headset branding) is in beta testing via a game called Giant Mini Paddle Golf. "Those in the test group will see a new 'Quests' icon in their Identity Panel, which will open the new quests board to show 6 quests (e.g. Get a Hole-in-one) and the rewards to be earned (e.g. Sea Captain Costume). With one click, travel to the world to get started," according to the description. Meta plans to roll it out to more users over time. 

While limited to just a single experience for now, it's easy to imagine Meta offering quests in other corners of Horizon World as a way to give Quest VR headset users more to do. Last month, the company said it planned to release 20 new Horizon experiences built by third-party studios, so perhaps the quests feature will be part of these. 

The Horizon Worlds user base was reportedly around 200,000 at the end of 2022, well short of the goal it originally set. Meta recently announced plans to open Horizon World up to children between 13-17 years old, prompting criticism from two US senators concerned over Meta's track record on protecting younger users. 

You may soon see ads on Instagram in places you didn't in the past. The Meta-owned app has started testing a couple of new ad placements meant to give businesses more ways to get discovered. One of those experimental placements puts ads in its search results. When you search for a particular term on the app — say "makeup" — posts marked "sponsored" will show up in the feed you can scroll through when you tap on any of the actual results. In its announcement, Instagram said it plans to roll out the placement globally in the coming months.

In addition, the app has teamed up with certain brands to test a format that would give businesses the power to remind or notify you of future events or launches. When you opt into the app's "reminder ads" for a particular event, you'll get notifications from Instagram one day before, 15 minutes before and when the event begins. These reminders will appear like any other Instagram notification and will show up on your lock screen. 

Instagram

Ads are Meta's lifeblood, and the introduction of new ways to earn from them comes as no surprise after a year that saw the company's quarterly revenue shrink for the first time. For the fourth quarter of 2022, for instance, the company reported an advertising revenue of $31.25 billion, down from $32.64 billion for the same quarter in 2021. Its year-on-year ad revenue was down, as well, from $114.93 billion in 2021 to $113.64 billion in 2022. Meta has been severely tightening belt in recent months and let more than 11,000 workers go in November in its first ever mass layoffs. Just a few days ago, company chief Mark Zuckerberg announced that Meta is laying off another 10,000 workers and will restructure its divisions in April and May.

LG is once again expanding its UltraGear line of gaming monitors with a new model. Before you ask: No, the company's latest is not a new OLED screen. But it is interesting for a few reasons. LG is marketing the 49GR85DC-B as a fast ultrawide with class-leading HDR capabilities.

The monitor features a 49-inch VA panel with a 32:9 aspect ratio, 5,120 by 1,440 resolution, 98.5 percent DCI-P3 coverage and an aggressive 1000R curve. It also features a 240Hz refresh rate, a claimed 1ms pixel response time and AMD FreeSync Premium. To top it all off, LG says the 49GR85DC-B is DisplayHDR 1000 certified, suggesting the panel is capable of peaking at an eye-searing 1000 nits of brightness and features some amount of local dimming.

At first glance, it’s a spec list that should make the 49GR85DC-B a homerun for ultrawide gaming fans, but there are a few things to note that may not make it as appealing as it seems. First, there’s the price. LG is asking $1,300 for the 49GR85DC-B. In 2023, that’s a lot for an LCD. Additionally, in my experience, VA panels are never as fast as manufacturers say they are, so don’t be surprised if LG’s latest doesn’t live up to the 1ms response time the company has listed. It's also worth mentioning few games support 32:9 resolutions.

The 49GR85DC-B is available to preorder starting today through LG’s website. If you decide to jump on this one early, the company will send you a complimentary UltraGear Gaming Pad. That's a $200 mouse mat that doubles as a USB hub and features RGB lighting.

Disentangling ourselves from TikTok is more complicated than simply banning the app, just ask the state of Maryland. According to a new report in The Wall Street Journal, it’s one of several states that used TikTok’s tracking pixel on a government website despite a statewide ban barring TikTok-related software from official devices and networks.

According to the report, Maryland was one of 27 states that had code for TikTok’s tracking pixel embedded in an official government website. While these types of tools are extremely common — tracking pixels help online advertisers target their ads — their use has also been widely criticized by privacy advocates.

In Maryland’s case, the TikTok pixels were reportedly found on a state-run COVID website and were related to an ad campaign from last year. Likewise, TikTok’s pixel was also found on a website run by Utah's Department of Workforce Services, which told The Wall Street Journal the pixel was used for an ad campaign targeting job seekers. Like Maryland, Utah has also banned TikTok from government devices.

The report underscores how, even with bans in place, governments are finding it difficult to disentangle themselves from TikTok completely. The company is currently grappling with the threat of a nationwide ban in the United States if parent company ByteDance doesn’t divest its stake in the service. CEO Shou Zi Chew is set to testify in a Congressional hearing on Thursday, when he will make the case that banning the app would hurt its 150 million American users.

Elsewhere, a new report in Forbes highlighted other issues that a nationwide ban may not fully resolve. According to the report, the personal data of TikTok users from India is still accessible to TikTok and ByteDance employees, despite the country banning the app in 2020. Forbes points out that this is likely due to the terms of India’s ban, which apparently “did not seem to call for deletion of app data that had already been captured and stored.”

Even so, it’s not the first time security experts have questioned whether it would ever be possible to “claw back” TikTok user data that’s already been collected by the company. In an odd way, that may make it a bit easier for TikTok to argue that an outright ban would be less effective than its multibillion-dollar plan to impose strict data controls and other measures meant to lock down US user data. That plan, known as Project Texas, has so far failed to persuade lawmakers and the Treasury Department officials involved in the years-long negotiations with TikTok.

Last month, Roblox outlined its vision for AI-assisted content creation, imagining a future where Generative AI could help users create code, 3D models and more with little more than text prompts. Now, it's taking its first steps toward allowing "every user on Roblox to be a creator" by launching its first AI tools: Code Assist and Material Generator, both in beta. 

Although neither tool is anywhere close to generating a playable Roblox experience from a text description, Head of Roblox Studio Stef Corazzatold an audience at GDC 2023 that they can "help automate basic coding tasks so you can focus on creative work." For now, that means being able to generate useful code snippets and object textures based on short prompts. Roblox's announcement for the tools offers a few examples, generating realistic textures for a "bright red rock canyon" and "stained glass," or producing several lines of functional code that will that make certain objects change color and self-destruct after a player interacts with them. 

Both of those functions might sound familiar to you if you've experimented AI chatbots — GPT-3 can already create functional code snippets based on prompts.  

Code Assist looks promising, but Roblox is careful to state it's imperfect, and may generate "incorrect" or "misleading" information. "It is still up to you to review, test, and determine if the code suggestion is contextually appropriate." Even so, Roblox's Corazza seems confident that this is the first step towards making every user on the platform a creator, suggesting it may only be a few years before these tools can generate fully playable, interactive 3D scenes from a simple prompt.

Google has made Bard more widely available to users in the US and the UK today, and I have been spending some time with the company's chatbot to see how its generative AI compares to ChatGPT and Bing AI. 

Like we saw in the screenshots Google provided with today's announcement, the interface here is very similar to Bing AI in that there is a wide text input at the bottom of the screen and a dialogue-based layout. But there are a few key differences between Google's and Microsoft's offerings. 

With Bing AI, you'll have to either hit Chat or scroll up from search results to get to the conversation page, whereas you don't have to do that for the Bard website. Microsoft has a broom icon to the left of the input bar to clear the slate and start a new topic, while Google has a column on the left with options for "Reset chat," "Bard Activity," "FAQ and "Help & Support." 

It's also worth noting the language Google painstakingly uses here. Once I navigated to the website, I was greeted with an alert reminding me that "Bard is an experiment." It asks users to remember two things: "Bard will not always get it right," and that "Bard will get better with your feedback."

Even after you click "Got it" and that reminder goes away, there's a line of fine print below the input field that states "Bard may display inaccurate or offensive information that doesn't represent Google's views." After the embarrassing blunders Bard has already made so far, it's understandable (and mildly funny) to see all these disclaimers. 

This story is developing, please refresh for updates.

The age of enterprise AI has come crashing down upon us in recent months. Public infatuation with ChatGPT since its release last November has opened the floodgates of corporate interest and set off an industry-wide land grab with every major tech entity vying to stake their claim in this burgeoning market by incorporating generative AI features into their existing products. Heavyweights including Google, Microsoft, Meta, and Baidu are already jockeying their Large Language Models (LLMs) for market dominance, while everybody else, from Adobe and AT&T to BMW and BYD, scrambles to find uses for the revolutionary technology.  

NVIDIA's newest cloud services offering, AI Foundations, will allow businesses lacking the time and money to develop their own models from scratch to "to build, refine and operate custom large language models and generative AI models that are trained with their own proprietary data and created for their unique domain-specific tasks."   

These models include NeMo, NVIDIA’s text-to-image generation engine and DALL-E 2 competitor; BioNemo, a drug and molecule discovery-focused fork of the NeMo model built for the medical research community; and Picasso, an AI capable of generating images, video and “3D applications… to supercharge productivity for creativity, design and digital simulation,” according to Tuesday’s release. Both flavors of NeMo are still in early access and Picasso remains in private preview despite Tuesday's news, so it'll be a minute before any of them are released to the wider public. NeMo and Picasso both operate on NVIDIA's new DGX Cloud platform and will eventually be accessible through an online portal.  

These enterprise-facing cloud-based services function as blank templates that companies can pour their own databases into to train on specifically. So while something like Google’s Bard AI is trained on (and will pull from) data from all over the internet to provide a generated response, NVIDIA’s AIs will allow companies to tailor a similarly-styled LLM to their own specific needs using their own proprietary data — think of ChatGPT but solely for one Pharma company's research division. The models can be trained with anywhere from 8 billion to 530 billion parameters, which is more than triple the 185 billion parameters GPT-3.5 provided.

Imagine StableDiffusion, but trained on Getty Images with Getty’s actual permission. NVIDIA announced such a system Tuesday built on the NeMo cloud service: a series of responsibly sourced text-to-image and text-to-video models, "trained on Getty Images' fully licensed assets," Tuesday’s press release read. "Getty Images will provide royalties to artists on any revenues generated from the models."

BioNeMo, uses the same technical underpinnings as NeMo itself, but is geared entirely towards drug and molecule discovery. Per Tuesday’s release, Bio NeMo, “enables researchers to fine-tune generative AI applications on their own proprietary data, and to run AI model inference directly in a web browser or through new cloud APIs that easily integrate into existing applications."

"BioNeMo is dramatically accelerating our approach to biologics discovery," Peter Grandsard, executive director of Biologics Therapeutic Discovery at Amgen said in a statement. "With it, we can pre-train large language models for molecular biology on Amgen’s proprietary data, enabling us to explore and develop therapeutic proteins for the next generation of medicine that will help patients." 

Six models will be available at launch including DeepMind’s AlphaFold2, Meta AI’s ESM2 and ESMFold predictive models ProtGPT-2, DiffDock and MoFlow. According to the companies, incorporating AI-based predictive models helped reduce the time to train “five custom models for molecule screening and optimization” using Amgen’s proprietary data on antibodies from the usual three months down to four weeks.

NVIDIA announced a similar partnership with Shutterstock as well. The photography site will use Picasso to generate 3D objects from text prompts as a new feature within Creative Flow, with plans to offer it on Turbosquid.com and NVIDIA’s forthcoming Omniverse platform.

“Our generative 3D partnership with NVIDIA will power the next generation of 3D contributor tools, greatly reducing the time it takes to create beautifully textured and structured 3D models,” Shutterstock CEO Paul Hennessy, said in the release. “This first of its kind partnership furthers our strategy of leveraging Shutterstock’s massive pool of metadata to bring new products, tools, and content to market. By combining our 3D content with NVIDIA’s foundation models, and utilizing our respective marketing and distribution platforms, we can capitalize on an extraordinarily large market opportunity.”

NVIDIA is also partnering with Adobe as part of the latter’s Content Authenticity Initiative, which seeks to enhance transparency and accountability within the generative AI training process. The CAI’s proposals include a “do not train” list, similar to robot.txt but for images and multimodal content, and persistent origination tags that will detail whether a piece is AI generated and from where. The two companies have also announced plans to incorporate many of Picasso’s features directly into Adobe’s suite of editing software including Photoshop, Premiere Pro and After Effects. 

When NVIDIA's founder and CEO Jensen Huang waxed poetic about artificial intelligence in the past, it mostly felt like marketing bluster, the sort of lofty rhetoric we've come to expect from an executive with a never-ending supply of leather jackets. But this year, following the hype around OpenAI's ChatGPT, Microsoft's revamped Bing and a slew of other competitors, NVIDIA's AI push finally seems to be leading somewhere.

The company's GTC (GPU Technology Conference) has always been a platform to promote its hardware for the AI world—now it's practically a celebration of how well-positioned NVIDIA is to take advantage of this moment. 

"We are at the iPhone moment for AI," Huang said during his GTC keynote this morning. He was quick to point out NVIDIA's role at the start of this AI wave: he personally brought a DGX AI supercomputer to OpenAI in 2016, hardware that was ultimately used to build ChatGPT. We've seen the DGX systems evolve over the years, but it's remained out of reach for many companies (the DGX A100 sold for $200,000 in 2020, which was half the price of its predecessor!). So what about everyone else?

That's where NVIDIA's new DGX Cloud comes in, an (obviously) online way to tap into the power of its AI supercomputers. Starting at a mere $36,999 a month for a single node, its meant to be a more flexible way for companies to scale up their AI needs. DGX Cloud can also work together with on-site DGX devices, since they're all controlled with NVIDIA's Base Command software.

NVIDIA says every DGX Cloud instance is powered by eight of its H100 or A100 systems with 60GB of VRAM, bringing the total amount of memory to 640GB across the node. There's high-performance storage, as you'd expect, as well as low-latency fabric that connects the systems together. That amount of power may make the cloud solution more tempting for existing DGX customers—why spend another $200,000 on a box, when you can do so much more for a lower monthly fee? DGX Cloud will be powered by Oracle's Cloud Infrastructure to start, but NVIDIA says it will expand to Microsoft Azure next quarter, as well as Google Cloud and other providers "soon."

Developing...

Microsoft's Bing AI chat can already be helpful for finding answers, but now it can help you produce fanciful pictures. The company has introduced a Bing Image Creator preview that adds OpenAI's DALL-E AI image generation to both Bing search and a sidebar in the Edge browser. You just have to ask the chatbot to create an image with either a direct description or a follow-up to a previous query. If you're wondering how to revamp your living room, you can ask Bing to draw some ideas based on your criteria.

Yes, Microsoft is aware of the potential for things to go awry. The company says it's applying "additional protections" beyond OpenAI's own. It will block you from creating potentially "harmful" images, the firm says. Microsoft also explicitly clarifies that images are AI-generated, including through watermarks.

Image Creator is available to a selection of Bing desktop and mobile users in preview and through a dedicated site. Edge users have access as well If you're part of the test group, you'll have to toggle the Creative mode to give the generator a try. Microsoft plans to bring the creative tool to Balanced and Precise mode users in the future, though, and plans to fine-tune the system's behavior in multi-step chats. While the technology only supports English, more languages are in the works.

More forms of AI are finding their way into Bing. The company is rolling out AI-based Knowledge Cards that now offer "dynamic" quick-glance info like charts and timelines. Stories, meanwhile, provide images and short videos linked to searched topics.

The DALL-E tool comes just as Adobe has unveiled plans to put generative AI in Photoshop, After Effects and Premiere Pro, while NVIDIA is launching a customizable cloud AI service that includes image creation. While Microsoft clearly isn't competing directly with Adobe or NVIDIA, it's joining a wave of tech giants that see AI image production as a valuable tool. Bing Image Creator might also provide a competitive edge over Google, which only just widened access to its Bard AI chat following a limited test.