Category: Technology

Watermarking ChatGPT and Other Generative AIs Could Help Protect Against Fraud and Misinformation

Shortly after rumors leaked of former President Donald Trump’s impending indictment, images purporting to show his arrest appeared online. These images looked like news photos, but they were fake. They were created by a generative artificial intelligence system.

Generative AI, in the form of image generators like DALL-E, Midjourney and Stable Diffusion, and text generators like Bard, ChatGPT, Chinchilla and LLaMA, has exploded in the public sphere. By combining clever machine-learning algorithms with billions of pieces of human-generated content, these systems can do anything from create an eerily realistic image from a caption, synthesize a speech in President Joe Biden’s voice, replace one person’s likeness with another in a video, or write a coherent 800-word op-ed from a title prompt.

This article was republished with permission from The Conversation, a news site dedicated to sharing ideas from academic experts. It represents the research-based findings and thoughts of, Hany Farid, Professor of Computer Science, University of California, Berkeley.

Even in these early days, generative AI is capable of creating highly realistic content. My colleague Sophie Nightingale and I found that the average person is unable to reliably distinguish an image of a real person from an AI-generated person. Although audio and video have not yet fully passed through the uncanny valley – images or models of people that are unsettling because they are close to but not quite realistic – they are likely to soon. When this happens, and it is all but guaranteed to, it will become increasingly easier to distort reality.

In this new world, it will be a snap to generate a video of a CEO saying her company’s profits are down 20%, which could lead to billions in market-share loss, or to generate a video of a world leader threatening military action, which could trigger a geopolitical crisis, or to insert the likeness of anyone into a sexually explicit video.

Advances in generative AI will soon mean that fake but visually convincing content will proliferate online, leading to an even messier information ecosystem. A secondary consequence is that detractors will be able to easily dismiss as fake actual video evidence of everything from police violence and human rights violations to a world leader burning top-secret documents.

As society stares down the barrel of what is almost certainly just the beginning of these advances in generative AI, there are reasonable and technologically feasible interventions that can be used to help mitigate these abuses. As a computer scientist who specializes in image forensics, I believe that a key method is watermarking.

Watermarks

There is a long history of marking documents and other items to prove their authenticity, indicate ownership and counter counterfeiting. Today, Getty Images, a massive image archive, adds a visible watermark to all digital images in their catalog. This allows customers to freely browse images while protecting Getty’s assets.

Imperceptible digital watermarks are also used for digital rights management. A watermark can be added to a digital image by, for example, tweaking every 10th image pixel so that its color (typically a number in the range 0 to 255) is even-valued. Because this pixel tweaking is so minor, the watermark is imperceptible. And, because this periodic pattern is unlikely to occur naturally, and can easily be verified, it can be used to verify an image’s provenance.

Even medium-resolution images contain millions of pixels, which means that additional information can be embedded into the watermark, including a unique identifier that encodes the generating software and a unique user ID. This same type of imperceptible watermark can be applied to audio and video.

The ideal watermark is one that is imperceptible and also resilient to simple manipulations like cropping, resizing, color adjustment and converting digital formats. Although the pixel color watermark example is not resilient because the color values can be changed, many watermarking strategies have been proposed that are robust – though not impervious – to attempts to remove them.

Watermarking and AI

These watermarks can be baked into the generative AI systems by watermarking all the training data, after which the generated content will contain the same watermark. This baked-in watermark is attractive because it means that generative AI tools can be open-sourced – as the image generator Stable Diffusion is – without concerns that a watermarking process could be removed from the image generator’s software. Stable Diffusion has a watermarking function, but because it’s open source, anyone can simply remove that part of the code.

OpenAI is experimenting with a system to watermark ChatGPT’s creations. Characters in a paragraph cannot, of course, be tweaked like a pixel value, so text watermarking takes on a different form.

Text-based generative AI is based on producing the next most-reasonable word in a sentence. For example, starting with the sentence fragment “an AI system can…,” ChatGPT will predict that the next word should be “learn,” “predict” or “understand.” Associated with each of these words is a probability corresponding to the likelihood of each word appearing next in the sentence. ChatGPT learned these probabilities from the large body of text it was trained on.

Generated text can be watermarked by secretly tagging a subset of words and then biasing the selection of a word to be a synonymous tagged word. For example, the tagged word “comprehend” can be used instead of “understand.” By periodically biasing word selection in this way, a body of text is watermarked based on a particular distribution of tagged words. This approach won’t work for short tweets but is generally effective with text of 800 or more words depending on the specific watermark details.

Generative AI systems can, and I believe should, watermark all their content, allowing for easier downstream identification and, if necessary, intervention. If the industry won’t do this voluntarily, lawmakers could pass regulation to enforce this rule. Unscrupulous people will, of course, not comply with these standards. But, if the major online gatekeepers – Apple and Google app stores, Amazon, Google, Microsoft cloud services and GitHub – enforce these rules by banning noncompliant software, the harm will be significantly reduced.

Signing Authentic Content

Tackling the problem from the other end, a similar approach could be adopted to authenticate original audiovisual recordings at the point of capture. A specialized camera app could cryptographically sign the recorded content as it’s recorded. There is no way to tamper with this signature without leaving evidence of the attempt. The signature is then stored on a centralized list of trusted signatures.

Although not applicable to text, audiovisual content can then be verified as human-generated. The Coalition for Content Provenance and Authentication (C2PA), a collaborative effort to create a standard for authenticating media, recently released an open specification to support this approach. With major institutions including Adobe, Microsoft, Intel, BBC and many others joining this effort, the C2PA is well positioned to produce effective and widely deployed authentication technology.

The combined signing and watermarking of human-generated and AI-generated content will not prevent all forms of abuse, but it will provide some measure of protection. Any safeguards will have to be continually adapted and refined as adversaries find novel ways to weaponize the latest technologies.

In the same way that society has been fighting a decadeslong battle against other cyber threats like spam, malware and phishing, we should prepare ourselves for an equally protracted battle to defend against various forms of abuse perpetrated using generative AI.

The Fear of Mass Unemployment Due to Artificial Intelligence and Robotics Is Unfounded

People are arguing over whether artificial intelligence (AI) and robotics will eliminate human employment. People seem to have an all-or-nothing belief that either the use of technology in the workplace will destroy human employment and purpose or it won’t affect it at all. The replacement of human jobs with robotics and AI is known as “technological unemployment.”

Although robotics can turn materials into economic goods in a fraction of the time it would take a human, in some cases using minimal human energy, some claim that AI and robotics will actually bring about increasing human employment. According to a 2020 Forbes projection, AI and robotics will be a strong creator of jobs and work for people across the globe in the near future. However, also in 2020, Daron Acemoglu and Pascual Restrepo published a study that projected negative job growth when AI and robotics replace human jobs, predicting significant job loss each time a robot replaces a human in the workplace. But two years later, an article in The Economist showed that many economists have backtracked on their projection of a high unemployment rate due to AI and robotics in the workplace. According to the 2022 Economist article, “Fears of a prolonged period of high unemployment did not come to pass. . . . The gloomy narrative, which says that an invasion of job-killing robots is just around the corner, has for decades had an extraordinary hold on the popular imagination.” So which scenario is correct?

Contrary to popular belief, no industrialized nation has ever completely replaced human energy with technology in the workplace. For instance, the steam shovel never put construction workers out of work; whether people want to work in construction is a different question. And bicycles did not become obsolete because of vehicle manufacturing: “Consumer spending on bicycles and accessories peaked at $8.3 billion in 2021,” according to an article from the World Economic Forum.

Do people generally think AI and robotics can run an economy without human involvement, energy, ingenuity, and cooperation? While AI and robotics have boosted economies, they cannot plan or run an economy or create technological unemployment worldwide. “Some countries are in better shape to join the AI competition than others,” according to the Carnegie Endowment for International Peace. Although an accurate statement, it misses the fact that productive economies adapt to technological changes better than nonproductive economies. Put another way, productive people are even more effective when they use technology. Firms using AI and robotics can lower production costs, lower prices, and stimulate demand; hence, employment grows if demand and therefore production increase. In the unlikely event that AI or robotic productive technology does not lower a firm’s prices and production costs, employment opportunities will decline in that industry, but employment will shift elsewhere, potentially expanding another industry’s capacity. This industry may then increase its use of AI and robotics, creating more employment opportunities there.

In the not-so-distant past, office administrators did not know how to use computers, but when the computer entered the workplace, it did not eliminate administrative employment as was initially predicted. Now here we are, walking around with minicomputers in our pants pockets. The introduction of the desktop computer did not eliminate human administrative workers—on the contrary, the computer has provided more employment since its introduction in the workplace. Employees and business owners, sometimes separated by time and space, use all sorts of technological devices, communicate with one another across vast networks, and can be increasingly productive.

I remember attending a retirement party held by a company where I worked decades ago. The retiring employee told us all a story about when the company brought in its first computer back in the late ’60s. The retiree recalled, “The boss said we were going to use computers instead of typewriters and paper to handle administrative tasks. The next day, her department went from a staff of thirty to a staff of five.” The day after the department installed computers, twenty-five people left the company to seek jobs elsewhere so they would not “have to learn and deal with them darn computers.”

People often become afraid of losing their jobs when firms introduce new technology, particularly technology that is able to replicate human tasks. However, mass unemployment due to technological innovation has never happened in any industrialized nation. The notion that AI will disemploy humans in the marketplace is unfounded. Mike Thomas noted in his article “Robots and AI Taking Over Jobs: What to Know about the Future of Jobs” that “artificial intelligence is poised to eliminate millions of current jobs—and create millions of new ones.” The social angst about the future of AI and robotics is reminiscent of the early nineteenth-century Luddites of England and their fear of replacement technology. Luddites, heavily employed in the textile industry, feared the weaving machine would take their jobs. They traveled throughout England breaking and vandalizing machines and new manufacturing technology because of their fear of technological unemployment. However, as the textile industry there became capitalized, employment in that industry actually grew. History tells us that technology drives the increase of work and jobs for humans, not the opposite.

We should look forward to unskilled and semiskilled workers’ upgrading from monotonous work because of AI and robotics. Of course, AI and robotics will have varying effects on different sectors; but as a whole, they are enablers and amplifiers of human work. As noted, the steam shovel did not disemploy construction workers. The taxi industry was not eliminated because of Uber’s technology; if anything, Uber’s new AI technology lowered the barriers of entry to the taxi industry. Musicians were not eliminated when music was digitized; instead, this innovation gave musicians larger platforms and audiences, allowing them to reach millions of people with the swipe of a screen. And dating apps running on AI have helped millions of people fall in love and live happily ever after.

About the Author

Raushan Gross is an Associate Professor of Business Management at Pfeiffer University. His works include Basic Entrepreneurship, Management and Strategy, and the e-book The Inspiring Life and Beneficial Impact of Entrepreneurs.

Integrating Humans with AI in Structural Design

David L. Chandler | MIT News Office

Modern fabrication tools such as 3D printers can make structural materials in shapes that would have been difficult or impossible using conventional tools. Meanwhile, new generative design systems can take great advantage of this flexibility to create innovative designs for parts of a new building, car, or virtually any other device.

But such “black box” automated systems often fall short of producing designs that are fully optimized for their purpose, such as providing the greatest strength in proportion to weight or minimizing the amount of material needed to support a given load. Fully manual design, on the other hand, is time-consuming and labor-intensive.

Now, researchers (MIT) have found a way to achieve some of the best of both of these approaches. They used an automated design system but stopped the process periodically to allow human engineers to evaluate the work in progress and make tweaks or adjustments before letting the computer resume its design process. Introducing a few of these iterations produced results that performed better than those designed by the automated system alone, and the process was completed more quickly compared to the fully manual approach.

The results are reported this week in the journal Structural and Multidisciplinary Optimization, in a paper by MIT doctoral student Dat Ha and assistant professor of civil and environmental engineering Josephine Carstensen.

The basic approach can be applied to a broad range of scales and applications, Carstensen explains, for the design of everything from biomedical devices to nanoscale materials to structural support members of a skyscraper. Already, automated design systems have found many applications. “If we can make things in a better way, if we can make whatever we want, why not make it better?” she asks.

“It’s a way to take advantage of how we can make things in much more complex ways than we could in the past,” says Ha, adding that automated design systems have already begun to be widely used over the last decade in automotive and aerospace industries, where reducing weight while maintaining structural strength is a key need.

“You can take a lot of weight out of components, and in these two industries, everything is driven by weight,” he says. In some cases, such as internal components that aren’t visible, appearance is irrelevant, but for other structures, aesthetics may be important as well. The new system makes it possible to optimize designs for visual as well as mechanical properties, and in such decisions, the human touch is essential.

As a demonstration of their process in action, the researchers designed a number of structural load-bearing beams, such as might be used in a building or a bridge. In their iterations, they saw that the design has an area that could fail prematurely, so they selected that feature and required the program to address it. The computer system then revised the design accordingly, removing the highlighted strut and strengthening some other struts to compensate, and leading to an improved final design.

The process, which they call Human-Informed Topology Optimization, begins by setting out the needed specifications — for example, a beam needs to be this length, supported on two points at its ends, and must support this much of a load. “As we’re seeing the structure evolve on the computer screen in response to initial specification,” Carstensen says, “we interrupt the design and ask the user to judge it. The user can select, say, ‘I’m not a fan of this region, I’d like you to beef up or beef down this feature size requirement.’ And then the algorithm takes into account the user input.”

While the result is not as ideal as what might be produced by a fully rigorous yet significantly slower design algorithm that considers the underlying physics, she says it can be much better than a result generated by a rapid automated design system alone. “You don’t get something that’s quite as good, but that was not necessarily the goal. What we can show is that instead of using several hours to get something, we can use 10 minutes and get something much better than where we started off.”

The system can be used to optimize a design based on any desired properties, not just strength and weight. For example, it can be used to minimize fracture or buckling, or to reduce stresses in the material by softening corners.

Carstensen says, “We’re not looking to replace the seven-hour solution. If you have all the time and all the resources in the world, obviously you can run these and it’s going to give you the best solution.” But for many situations, such as designing replacement parts for equipment in a war zone or a disaster-relief area with limited computational power available, “then this kind of solution that catered directly to your needs would prevail.”

Similarly, for smaller companies manufacturing equipment in essentially “mom and pop” businesses, such a simplified system might be just the ticket. The new system they developed is not only simple and efficient to run on smaller computers, but it also requires far less training to produce useful results, Carstensen says. A basic two-dimensional version of the software, suitable for designing basic beams and structural parts, is freely available now online, she says, as the team continues to develop a full 3D version.

“The potential applications of Prof Carstensen’s research and tools are quite extraordinary,” says Christian Málaga-Chuquitaype, a professor of civil and environmental engineering at Imperial College London, who was not associated with this work. “With this work, her group is paving the way toward a truly synergistic human-machine design interaction.”

“By integrating engineering ‘intuition’ (or engineering ‘judgement’) into a rigorous yet computationally efficient topology optimization process, the human engineer is offered the possibility of guiding the creation of optimal structural configurations in a way that was not available to us before,” he adds. “Her findings have the potential to change the way engineers tackle ‘day-to-day’ design tasks.”

Reprinted with permission from MIT News ( http://news.mit.edu/ )