It’s astounding how quickly artificial intelligence seems to be developing. Many of us believed that AI wouldn’t be useful on its own for a few years before the end of last year. Then GPT-3-backed ChatGPT arrived, swiftly followed by GPT-4, and all of a sudden, artificial intelligence that can boost productivity is all around us.
The most intriguing is the DeepBrain AI model for video synthesis, which can produce a functional digital twin of a person that can increasingly perform tasks that the person it represents would normally have to perform.
This week, let’s discuss the rise of the practical human digital twin. My pick for Product of the Week is Lenovo’s Motorola ThinkPhone, a smartphone that corrects one of the major errors made by the computer industry in its reaction to Apple’s iPhone.
AI models for Video Synthesis
The first human digital twin I’ve looked at that can complete tasks exactly as a human would have is the DeepBrain AI, which was initially concentrated on news anchor talent.
A vast database of information about the mannerisms, speech patterns, and physical attributes of human presenters is built up by letting the digital twin learn from a variety of news videos.
If a script is fed to the resulting digital twin, it will act like a news presenter. The news service pays the talent each time it uses this avatar, which is mostly for breaking news or brief format teases for upcoming live programming. The digital twin does not take the place of the human talent.
An AI-generated avatar is occasionally noted by the news services as being watched by the public. When this technology is already in use, both the talent and the news audience have embraced it.
The news talent is cool with this because they get paid when the avatar is used, and using the avatar saves them from having to travel to a studio in order to tape or record live short segments, updates, or announcements, which typically require them to return to a studio.
Thus, their income is unaffected, and the avatar enhances human labor by lightening the workload and associated stress, all of which should increase employee satisfaction.
Use cases for avatars
Initially, this technology was mostly used for video news programs where the newscaster and avatar could not be distinguished (as in Asia), but there are other applications as well.
Banks are starting to use virtual kiosks where customers can converse with AIs pretending to be humans.
training videos that are one-way and follow a script.
A virtual hotel concierge who can help with bookings for shows or restaurants, for example.
interactive films that evoke the sensation of speaking with a real person once more.
A number of these avatars were artificial intelligence-generated, incredibly lifelike pictures rather than actual people.
Cost benefits are substantial because producing some of this short-form content with a live subject normally costs about $4,000, but using a computer-generated avatar only costs about $100.
Without the need for rehearsal, the avatar can work directly from the script without interruptions from illnesses, conflicts, or behavioral issues that are usually present in live performances.
The avatar is controlled by other AI, such as ChatGPT or IBM’s Watson, because it operates through text. This allows for the creation of interactive content that is so human-like that it has the potential to trick a large number of users into believing they are speaking with a real person.
Celebrity advocacy has also been investigated
Because there is very little opportunity for people to communicate with celebrities when they are used as advocates, celebrities don’t scale well. However, in addition to the client’s present or potential clients, a celebrity avatar could communicate with fans on a large scale.
As mentioned previously, this technology can scan real people as well as creatively created characters. This could potentially address some of the issues that businesses like Jack in the Box, Kentucky Fried Chicken, and McDonald’s (Ronald McDonald) had with using live actors. These businesses would frequently switch out the actors to avoid making them too attached to their roles.
Once the human element is eliminated, the company owns the virtual actor produced by a digitally generated avatar, and the issues that were originally caused by humans vanish.
What’s ahead for virtual and human digital twins?
DeepBrain’s current focus is mostly on improving people, not replacing them. But rather than using an enhancement model, their usage of fully synthetic avatars disconnected from humans is more of a replacement. Nothing stops the technology from eventually advancing to long-form productions like TV series and movies, even though their initial focus is on short-form content.
With more training and technological advancements, the avatars’ hyper-realistic qualities will get better—even in extended form. Similar to how products like Dalle-E create art from composite images, the virtual actors will eventually blend in with real people despite being amalgams of those people.
There’s a big chance that technology like this will upend the media industry, especially since using virtually created content has enormous cost advantages over live action. It’s not only about the talent that will be lost in the future. It might also be possible to waive the talent’s entire studio fee.
This can be placed on the roadmap for fully photorealistic films and television series produced entirely by AIs that adapt to user preferences, as GPT-4 is already producing intriguing scripts and stories.
Ultimately, this technology in conjunction with generative AI may produce personalized videos at scale, maybe featuring you and your family as the primary characters (of course, with your consent) instead of everyone else watching the same TV series and movies.
Granted, you could then post those videos on social media to attract the attention of people who are curious about what other people are producing. This could result in the creation of a vast amount of original content that services would have to review and offer to a global audience of prospective clients.
Finishing Up
It would be revolutionary if DeepBrain AI could design your twin to perform some of your tasks. However, it will drastically disrupt all forms of entertainment when we can do the same with synthesized images, especially for long-form content. This is already well known to the pornographic industry, for example. The majority of users don’t seem to care that Reddit content has been doing this for a while.
The possibilities are endless and point to a degree of disruption we are only now starting to witness. Examples include in-game non-player characters (NPCs) that behave like actual people and the ability to create entire virtual sports teams with lifelike representations of real, imagined, or even deceased players.
To put it briefly, the challenge of distinguishing between reality and fakery is already formidable and is only expected to worsen. This may prove beneficial for entertainment purposes, but it might have far more detrimental effects on our capacity to perceive the truth. Not even close to being prepared for that.
Lenovo ThinkPhone manufactured by Motorola
The computer industry made a major error when it completely shifted its focus to the iPhone. Ironically, the industry first disregarded smartphones that were geared toward consumers. Then, instead of opposing that trend on the basis of its merits, it tried to change course and use the iPhone to propel Apple from a market laggard to the leader of the pack almost instantly.
There had been previous instances of it. A few decades prior, IBM made an attempt to switch to client/server computing and nearly gave up on the mainframe, which caused it to go from being the undisputed market leader to virtually going out of business in a short amount of time.
In terms of smartphones with a business focus, there is currently an underserved market. I had a conversation with the former CEO of Bank of America, who expressed his disapproval of his employees using cell phones. The lack of a BlackBerry-like device that was both secure and business-oriented was a source of frustration for him.
Lenovo has now addressed that with its renowned ThinkPhone.
The ThinkPhone, which was developed by IBM, is akin to an iPhone in size but boasts a number of exclusive business-focused features. It builds upon the highly business-focused ThinkPad brand.
These attributes consist of:
Instant Connect: Using Wi-Fi, the phone and computer automatically identify one another when they are close.
Unified Clipboard: Copy text, recently taken pictures, scanned documents, and videos between devices without any interruptions by pasting them into any app on the other device.
Unified Notifications: The Windows Action Center instantly displays phone notifications. The PC screen of the associated phone app opens automatically when a notification is clicked.
File Drop: Drag and drop files between pc\. and ThinkPhone with ease.
App Streaming: Use a pc\. to open any Android application straight away.
Advanced Webcam: Use the ThinkPhone’s strong cameras and AI capabilities to your advantage by using it as a smooth webcam for all of your video calls. You already have a better camera on your smartphone, so why buy a separate webcam?
Instant Hotspot: Utilize the ThinkPhone’s 5G connectivity by connecting to the internet with just one click from the PC. This could reduce the requirement and expense for WAN capability in your piece, which is really significant.
Like the ThinkPad, the ThinkPhone is tested to Military Standard 810H (MIL-STD-810H) and wrapped in security due to its use of aircraft-grade aluminum, aramid fiber (used in bulletproof vests), and Gorilla Glass’s most durable solution, Victus.
The ThinkPhone can withstand submersion for up to thirty minutes at a depth of 1.5 meters. It even features a red button that can be used to open a crucial app. Since I usually need to access the camera quickly, I usually choose it. The Microsoft Teams Walkie Talkie app allows for the reintroduction of push-to-talk for law enforcement, security, and other applications where prompt communication is essential.
Because ThinkPhone is built to support remote management, it can be centrally configured and managed to guarantee security and prevent misuse, which is a common need for a computing device intended for business use. With its exclusive Moto Secure processor, ThinkPhone keeps cryptographic keys, PINs, and passwords separate and in an environment that is difficult for hackers to access.
The ThinkPhone comes with a special, incredibly compact 68W universal charger that can be used to power most USB-C devices, including workstations and gaming computers, and charge the phone in a matter of minutes.
At last, the phone has a superb 50 MP camera that should suffice for the majority of photo needs, be it capturing a special occasion or providing high-quality documentation for an insurance investigator or someone else.
My Product of the Week is the ThinkPhone, which replaces the gap in business phones left by the withdrawal of Palm and BlackBerry from the market.