... Inworld offers a Character Engine that constructs virtual beings out of text descriptions. Users can detail the appearance, behavior, background, and more in great detail, and the generative AI engine will synthesize the virtual being described, relying on general knowledge databases and proprietary sources the user can include. Though Inworld has been employed for education and marketing, the chief focus of the company is entertainment, especially video games. The multimodal characters rely on a mix of machine learning models that Inworld describes as going beyond ChatGPT or other conversational AIs relying on large language models (LLMs). ...
Inworld’s raise continues the startup’s rapid-fire rollout of new features and partnerships since coming out of stealth in 2021. Last year, the startup debuted Inworld Studio, raised $60 million in two funding rounds, and received an Epic MegaGrant from Unity Creator Epic Games. Inworld also joined and graduated from the Disney Accelerator program, where it demonstrated a prototype ‘Droid Maker’ for designing the look and personality of an interactive droid from the Star Wars universe. ...
The real threat from the ever-increasing use of AI is twofold. On the one hand, it’s a fantastic tool for frauds and cheats. On the other hand, it can spit out false information as easily as fact. Either way, the truth is the victim. ...
AI Will Not Keep Your Secrets
Not everything that’s true needs to be public, and thereby hangs another tale of AI woe. Coders love using AI because it can spit out code on demand, debug or transform existing code, and even write comments. But in most cases, whatever you put into the AI becomes part of its training. A few developers at Samsung got into hot water over feeding an AI proprietary code for debugging. AI may not take your job, but misusing it might get you fired. ...
... When you use AI for research (and you almost certainly will), don’t take results at face value. Go to the original sources. I much appreciate the fact that Bing AI footnotes its results, so finding sources is easy. ...
... Magic Leap’s latest solution aims to bypass traditional augmented dashboards with its immersive AR goggles. With a wide field of view (FoV), the head-mounted display (HMD) offers drivers a trove of features. ...
The unprecedented concept car aims to combine AR overlays with an intuitive HUD, complete with a 70-degree FoV. This offers instantaneous data on road conditions and streamlines components for the vehicle by eliminating dashboards or satnav systems.
Offering similar vehicle feedback as the Vision 111, Magic Leap’s solution also hopes to increase road safety and scale up the future potential of augmented automotive interfaces. This additionally offers drivers low latency, real-time hand and eye-tracking solutions. ...
... In the “Migraine Experience”, VR was used to educate the general public on the severity of migraines – they’re “more than just a headache”. People who don’t usually get migraines had the opportunity to experience the common symptoms first-hand – including, blurred vision, disorientation, and sensitivity to sound and light. With their new medication, GlaxoSmithKline also made sure to establish themselves as the solution to the pervasive problem of migraines (currently affecting 39 million people in the United States). Participant reactions were also filmed and compiled into an impactful marketing video that’s been viewed on social media over twelve million times. ...
... There’s a big problem, though: identifying material that was created by artificial intelligence is a massive technical challenge. The best options currently available—detection tools powered by AI, and watermarking—are inconsistent, impermanent, and sometimes inaccurate. (In fact, just this week OpenAI shuttered its own AI-detecting tool because of high error rates.) ...
But another approach has been attracting attention lately: C2PA. Launched two years ago, it’s an open-source internet protocol that relies on cryptography to encode details about the origins of a piece of content, or what technologists refer to as “provenance” information.
The developers of C2PA often compare the protocol to a nutrition label, but one that says where content came from and who—or what—created it. ...
The major media platform Shutterstock has joined as a member and announced its intention to use the protocol to label all its AI-generated content, including its DALL-E-powered AI image generator. ...
What is C2PA and how is it being used?
Microsoft, Intel, Adobe, and other major tech companies started working on C2PA in February 2021, hoping to create a universal internet protocol that would allow content creators to opt in to labeling their visual and audio content with information about where it came from. (At least for the moment, this does not apply to text-based posts.)
Crucially, the project is designed to be adaptable and functional across the internet, and the base computer code is accessible and free to anyone. ...
More specifically, it works by encoding provenance information through a set of hashes that cryptographically bind to each pixel, says Jenks, who also leads Microsoft’s work on C2PA. ...
C2PA offers some critical benefits over AI detection systems, which use AI to spot AI-generated content and can in turn learn to get better at evading detection. It’s also a more standardized and, in some instances, more easily viewable system than watermarking, the other prominent technique used to identify AI-generated content. The protocol can work alongside watermarking and AI detection tools as well, says Jenks. ...
That said, provenance information is far from a fix-all solution. C2PA is not legally binding, and without required internet-wide adoption of the standard, unlabeled AI-generated content will exist, says Siwei Lyu, a director of the Center for Information Integrity and professor at the University at Buffalo in New York. “The lack of over-board binding power makes intrinsic loopholes in this effort,” he says, though he emphasizes that the project is nevertheless important. ...
What’s more, since C2PA relies on creators to opt in, the protocol doesn’t really address the problem of bad actors using AI-generated content. And it’s not yet clear just how helpful the provision of metadata will be when it comes to media fluency of the public. Provenance labels do not necessarily mention whether the content is true or accurate. ...
Ultimately, the coalition’s most significant challenge may be encouraging widespread adoption across the internet ecosystem, especially by social media platforms. ...
At the end of each day, one half of each couple is shown videos of their partner kissing or behaving sexually with the singles.
Watching the reactions of the individuals who are seeing their partners cheating on them and breaking their already fragile trust makes for gutting viewing - but it’s not real.
In fact, it’s not until after the couples are played the traumatic videos - many of them visibly devastated - that they’re told the films could be fake.
Host Raquel Sánchez Silva explains to the participants that the scenarios their partners are being unfaithful in are mostly false, altered with deep fake technology.
With that, the game is afoot. ...
It turns out there’s an €100,000 prize at stake but it isn’t awarded on the basis of fidelity or trust. Instead, the winners will be the couple who are able to decipher the largest number of videos as either real or deep fake. ...
The American director Jay Scheib was looking at a bank of monitors inside the Bayreuth Festival Theater on a recent afternoon.
He was rehearsing his new production of Wagner’s “Parsifal,”which opens the storied Bayreuth Festival on Tuesday, and as performers circled a large metallic monolith onstage, the screens showed three-dimensional flowers floating through blank space — psychedelic animations that will come to life for audience members who see them with augmented-reality glasses. ...
Scheib’s production is one of the most ambitious, and high-profile, attempts to incorporate augmented reality into opera performance. But it also caps months of tumult at Bayreuth, after plans to outfit nearly 2,000 audience members with the glasses for each performance were downscaled because of an apparent money dispute between the festival’s artistic and financial leadership. The compromise, in which only 330 attendees will be provided with the glasses to experience the production’s signature flourishes, has left many fuming, and concerned that internal conflicts at one of the most important events in opera were undermining its relevance. ...
The use of the technology, Scheib said, was in keeping with Wagner’s own way of approaching opera. “He carried out so many innovations, with lighting and architecture,” he added. “Ultimately, he wanted the theater to completely disappear.” ...
Overall, the White House’s deal with A.I. companies seems more symbolic than substantive. There is no enforcement mechanism to make sure companies follow these commitments, and many of them reflect precautions that A.I. companies are already taking.
Still, it’s a reasonable first step. And agreeing to follow these rules shows that the A.I. companies have learned from the failures of earlier tech companies, which waited to engage with the government until they got into trouble. In Washington, at least where tech regulation is concerned, it pays to show up early.
Commitment 1: The companies commit to internal and external security testing of their A.I. systems before their release.
Commitment 2: The companies commit to sharing information across the industry and with governments, civil society and academia on managing A.I. risks.
Commitment 3: The companies commit to investing in cybersecurity and insider-threat safeguards to protect proprietary and unreleased model weights.
Commitment 4: The companies commit to facilitating third-party discovery and reporting of vulnerabilities in their A.I. systems.
Commitment 5: The companies commit to developing robust technical mechanisms to ensure that users know when content is A.I. generated, such as a watermarking system.
Commitment 6: The companies commit to publicly reporting their A.I. systems’ capabilities, limitations, and areas of appropriate and inappropriate use.
Commitment 7: The companies commit to prioritizing research on the societal risks that A.I. systems can pose, including on avoiding harmful bias and discrimination and protecting privacy.
Commitment 8: The companies commit to develop and deploy advanced A.I. systems to help address society’s greatest challenges.
On July 12, 2023, the U.S. Senate Committee on the Judiciary’s Subcommittee on Intellectual Property held its second hearing on artificial intelligence (AI) and intellectual property (IP). The first hearing, held on June 7, 2023, focused on AI’s implications for patent law, while this second hearing centered on copyright issues. ...
Front and center was the thorny issue of training generative AI models. ...
Accordingly, the senators focused their questioning on mainly five topics, providing a glimpse of what potential federal legislation could look like:
Need for New Federal Legislation?
Licensing for Training, and Questions of Opt-Out and Opt-In Policies
Fair Use
Comparative International Standards
Consumer Disclosures
While the senators certainly seemed eager in their questions to understand the AI landscape and what Congress can do to protect IP and creators’ rights, how serious Congress is about enacting federal legislation to protect the creative industry and regulate generative AI remains to be seen. It also remains an open question how these new proposals might interact with existing enforcement tools under copyright law, the Lanham Act, and state laws. ...
