New research: “ Deception abilities emerged in large language models “:

Abstract: Large language models (LLMs) are currently at the forefront of intertwining AI systems with human communication and everyday life. Thus, aligning them with human values is of great importance. However, given the steady increase in reasoning abilities, future LLMs are under suspicion of becoming able to deceive human operators and utilizing this ability to bypass monitoring efforts. As a prerequisite to this, LLMs need to possess a conceptual understanding of deception strategies. This study reveals that such strategies emerged in state-of-the-art LLMs, but were nonexistent in earlier LLMs. We conduct a series of experiments showing that state-of-the-art LLMs are able to understand and induce false beliefs in other agents, that their performance in complex deception scenarios can be amplified utilizing chain-of-thought reasoning, and that eliciting Machiavellianism in LLMs can trigger misaligned deceptive behavior. GPT-4, for instance, exhibits deceptive behavior in simple test scenarios 99.16% of the time (P < 0.001). In complex second-order deception test scenarios where the aim is to mislead someone who expects to be deceived, GPT-4 resorts to deceptive behavior 71.46% of the time (P < 0.001) when augmented with chain-of-thought reasoning. In sum, revealing hitherto unknown machine behavior in LLMs, our study contributes to the nascent field of machine psychology.

Microsoft recently caught state-backed hackers using its generative AI tools to help with their attacks. In the security community, the immediate questions weren’t about how hackers were using the tools (that was utterly predictable), but about how Microsoft figured it out. The natural conclusion was that Microsoft was spying on its AI users, looking for harmful hackers at work.

Some pushed back at characterizing Microsoft’s actions as “spying.” Of course cloud service providers monitor what users are doing. And because we expect Microsoft to be doing something like this, it’s not fair to call it spying.

We see this argument as an example of our shifting collective expectations of privacy. To understand what’s happening, we can learn from an unlikely source: fish.

In the mid-20th century, scientists began noticing that the number of fish in the ocean—so vast as to underlie the phrase “There are plenty of fish in the sea”—had started declining rapidly due to overfishing. They had already seen a similar decline in whale populations, when the post-WWII whaling industry nearly drove many species extinct. In whaling and later in commercial fishing, new technology made it easier to find and catch marine creatures in ever greater numbers. Ecologists, specifically those working in fisheries management, began studying how and when certain fish populations had gone into serious decline.

One scientist, Daniel Pauly , realized that researchers studying fish populations were making a major error when trying to determine acceptable catch size. It wasn’t that scientists didn’t recognize the declining fish populations. It was just that they didn’t realize how significant the decline was. Pauly noted that each generation of scientists had a different baseline to which they compared the current statistics, and that each generation’s baseline was lower than that of the previous one.

What seems normal to us in the security community is whatever was commonplace at the beginning of our careers .

Pauly called this “ shifting baseline syndrome ” in a 1995 paper. The baseline most scientists used was the one that was normal when they began their research careers. By that measure, each subsequent decline wasn’t significant, but the cumulative decline was devastating. Each generation of researchers came of age in a new ecological and technological environment, inadvertently masking an exponential decline.

Pauly’s insights came too late to help those managing some fisheries. The ocean suffered catastrophes such as the complete collapse of the Northwest Atlantic cod population in the 1990s.

Internet surveillance, and the resultant loss of privacy, is following the same trajectory. Just as certain fish populations in the world’s oceans have fallen 80 percent, from previously having fallen 80 percent, from previously having fallen 80 percent (ad infinitum), our expectations of privacy have similarly fallen precipitously. The pervasive nature of modern technology makes surveillance easier than ever before, while each successive generation of the public is accustomed to the privacy status quo of their youth. What seems normal to us in the security community is whatever was commonplace at the beginning of our careers.

Historically, people controlled their computers, and software was standalone. The always-connected cloud-deployment model of software and services flipped the script. Most apps and services are designed to be always-online, feeding usage information back to the company. A consequence of this modern deployment model is that everyone—cynical tech folks and even ordinary users—expects that what you do with modern tech isn’t private. But that’s because the baseline has shifted.

AI chatbots are the latest incarnation of this phenomenon: They produce output in response to your input, but behind the scenes there’s a complex cloud-based system keeping track of that input—both to improve the service and to sell you ads .

Shifting baselines are at the heart of our collective loss of privacy. The U.S. Supreme Court has long held that our right to privacy depends on whether we have a reasonable expectation of privacy . But expectation is a slippery thing: It’s subject to shifting baselines.

The question remains: What now? Fisheries scientists, armed with knowledge of shifting-baseline syndrome, now look at the big picture. They no longer consider relative measures, such as comparing this decade with the last decade. Instead, they take a holistic, ecosystem-wide perspective to see what a healthy marine ecosystem and thus sustainable catch should look like. They then turn these scientifically derived sustainable-catch figures into limits to be codified by regulators.

In privacy and security, we need to do the same. Instead of comparing to a shifting baseline, we need to step back and look at what a healthy technological ecosystem would look like: one that respects people’s privacy rights while also allowing companies to recoup costs for services they provide. Ultimately, as with fisheries, we need to take a big-picture perspective and be aware of shifting baselines. A scientifically informed and democratic regulatory process is required to preserve a heritage—whether it be the ocean or the Internet—for the next generation.

This essay was written with Barath Raghavan, and previously appeared in IEEE Spectrum .

Read 24 remaining paragraphs | Comments

As India concluded the world’s largest election on June 5, 2024, with over 640 million votes counted, observers could assess how the various parties and factions used artificial intelligence technologies—and what lessons that holds for the rest of the world.

The campaigns made extensive use of AI, including deepfake impersonations of candidates, celebrities and dead politicians. By some estimates, millions of Indian voters viewed deepfakes.

But, despite fears of widespread disinformation, for the most part the campaigns, candidates and activists used AI constructively in the election. They used AI for typical political activities, including mudslinging, but primarily to better connect with voters.

Deepfakes without the deception

Political parties in India spent an estimated US$50 million on authorized AI-generated content for targeted communication with their constituencies this election cycle. And it was largely successful.

Indian political strategists have long recognized the influence of personality and emotion on their constituents, and they started using AI to bolster their messaging. Young and upcoming AI companies like The Indian Deepfaker , which started out serving the entertainment industry, quickly responded to this growing demand for AI-generated campaign material.

In January, Muthuvel Karunanidhi, former chief minister of the southern state of Tamil Nadu for two decades, appeared via video at his party’s youth wing conference. He wore his signature yellow scarf, white shirt, dark glasses and had his familiar stance—head slightly bent sideways. But Karunanidhi died in 2018. His party authorized the deepfake.

In February, the All-India Anna Dravidian Progressive Federation party’s official X account posted an audio clip of Jayaram Jayalalithaa, the iconic superstar of Tamil politics colloquially called “Amma” or “Mother.” Jayalalithaa died in 2016.

Meanwhile, voters received calls from their local representatives to discuss local issues—except the leader on the other end of the phone was an AI impersonation. Bhartiya Janta Party (BJP) workers like Shakti Singh Rathore have been frequenting AI startups to send personalized videos to specific voters about the government benefits they received and asking for their vote over WhatsApp.

Multilingual boost

Deepfakes were not the only manifestation of AI in the Indian elections. Long before the election began, Indian Prime Minister Narendra Modi addressed a tightly packed crowd celebrating links between the state of Tamil Nadu in the south of India and the city of Varanasi in the northern state of Uttar Pradesh. Instructing his audience to put on earphones, Modi proudly announced the launch of his “new AI technology” as his Hindi speech was translated to Tamil in real time.

In a country with 22 official languages and almost 780 unofficial recorded languages , the BJP adopted AI tools to make Modi’s personality accessible to voters in regions where Hindi is not easily understood. Since 2022, Modi and his BJP have been using the AI-powered tool Bhashini , embedded in the NaMo mobile app , to translate Modi’s speeches with voiceovers in Telugu, Tamil, Malayalam, Kannada, Odia, Bengali, Marathi and Punjabi.

As part of their demos, some AI companies circulated their own viral versions of Modi’s famous monthly radio show “Mann Ki Baat,” which loosely translates to “From the Heart,” which they voice cloned to regional languages.

Adversarial uses

Indian political parties doubled down on online trolling, using AI to augment their ongoing meme wars. Early in the election season, the Indian National Congress released a short clip to its 6 million followers on Instagram, taking the title track from a new Hindi music album named “Chor” (thief). The video grafted Modi’s digital likeness onto the lead singer and cloned his voice with reworked lyrics critiquing his close ties to Indian business tycoons.

The BJP retaliated with its own video , on its 7-million-follower Instagram account, featuring a supercut of Modi campaigning on the streets, mixed with clips of his supporters but set to unique music. It was an old patriotic Hindi song sung by famous singer Mahendra Kapoor , who passed away in 2008 but was resurrected with AI voice cloning.

Modi himself quote-tweeted an AI-created video of him dancing—a common meme that alters footage of rapper Lil Yachty on stage—commenting “such creativity in peak poll season is truly a delight.”

In some cases, the violent rhetoric in Modi’s campaign that put Muslims at risk and incited violence was conveyed using generative AI tools, but the harm can be traced back to the hateful rhetoric itself and not necessarily the AI tools used to spread it.

The Indian experience

India is an early adopter, and the country’s experiments with AI serve as an illustration of what the rest of the world can expect in future elections. The technology’s ability to produce nonconsensual deepfakes of anyone can make it harder to tell truth from fiction, but its consensual uses are likely to make democracy more accessible.

The Indian election’s embrace of AI that began with entertainment, political meme wars, emotional appeals to people, resurrected politicians and persuasion through personalized phone calls to voters has opened a pathway for the role of AI in participatory democracy.

The surprise outcome of the election, with the BJP’s failure to win its predicted parliamentary majority, and India’s return to a deeply competitive political system especially highlights the possibility for AI to have a positive role in deliberative democracy and representative governance.

Lessons for the world’s democracies

It’s a goal of any political party or candidate in a democracy to have more targeted touch points with their constituents. The Indian elections have shown a unique attempt at using AI for more individualized communication across linguistically and ethnically diverse constituencies, and making their messages more accessible, especially to rural, low-income populations.

AI and the future of participatory democracy could make constituent communication not just personalized but also a dialogue, so voters can share their demands and experiences directly with their representatives—at speed and scale.

India can be an example of taking its recent fluency in AI-assisted party-to-people communications and moving it beyond politics. The government is already using these platforms to provide government services to citizens in their native languages.

If used safely and ethically, this technology could be an opportunity for a new era in representative governance, especially for the needs and experiences of people in rural areas to reach Parliament.

This essay was written with Vandinika Shukla and previously appeared in The Conversation .

Read 9 remaining paragraphs | Comments

The debate over professionalizing software engineers is decades old. (The basic idea is that, like lawyers and architects, there should be some professional licensing requirement for software engineers.) Here’s a law journal article recommending the same idea for AI engineers.

This Article proposes another way: professionalizing AI engineering. Require AI engineers to obtain licenses to build commercial AI products, push them to collaborate on scientifically-supported, domain-specific technical standards, and charge them with policing themselves. This Article’s proposal addresses AI harms at their inception, influencing the very engineering decisions that give rise to them in the first place. By wresting control over information and system design away from companies and handing it to AI engineers, professionalization engenders trustworthy AI by design. Beyond recommending the specific policy solution of professionalization, this Article seeks to shift the discourse on AI away from an emphasis on light-touch, ex post solutions that address already-created products to a greater focus on ex ante controls that precede AI development. We’ve used this playbook before in fields requiring a high level of expertise where a duty to the public welfare must trump business motivations. What if, like doctors, AI engineers also vowed to do no harm?

I have mixed feelings about the idea. I can see the appeal, but it never seemed feasible. I’m not sure it’s feasible today.

This mini-essay was my contribution to a round table on Power and Governance in the Age of AI .  It’s nothing I haven’t said here before, but for anyone who hasn’t read my longer essays on the topic, it’s a shorter introduction.

The increasingly centralized control of AI is an ominous sign . When tech billionaires and corporations steer AI, we get AI that tends to reflect the interests of tech billionaires and corporations, instead of the public. Given how transformative this technology will be for the world, this is a problem.

To benefit society as a whole we need an AI public option —not to replace corporate AI but to serve as a counterbalance —as well as stronger democratic institutions to govern all of AI. Like public roads and the federal postal system, a public AI option could guarantee universal access to this transformative technology and set an implicit standard that private services must surpass to compete.

Widely available public models and computing infrastructure would yield numerous benefits to the United States and to broader society. They would provide a mechanism for public input and oversight on the critical ethical questions facing AI development, such as whether and how to incorporate copyrighted works in model training, how to distribute access to private users when demand could outstrip cloud computing capacity, and how to license access for sensitive applications ranging from policing to medical use. This would serve as an open platform for innovation, on top of which researchers and small businesses—as well as mega-corporations—could build applications and experiment. Administered by a transparent and accountable agency, a public AI would offer greater guarantees about the availability, equitability, and sustainability of AI technology for all of society than would exclusively private AI development.

Federally funded foundation AI models would be provided as a public service, similar to a health care public option. They would not eliminate opportunities for private foundation models, but they could offer a baseline of price, quality, and ethical development practices that corporate players would have to match or exceed to compete.

The key piece of the ecosystem the government would dictate when creating an AI public option would be the design decisions involved in training and deploying AI foundation models. This is the area where transparency, political oversight, and public participation can, in principle, guarantee more democratically-aligned outcomes than an unregulated private market.

The need for such competent and faithful administration is not unique to AI, and it is not a problem we can look to AI to solve. Serious policymakers from both sides of the aisle should recognize the imperative for public-interested leaders to wrest control of the future of AI from unaccountable corporate titans. We do not need to reinvent our democracy for AI, but we do need to renovate and reinvigorate it to offer an effective alternative to corporate control that could erode our democracy.

Researchers ran a global prompt hacking competition, and have documented the results in a paper that both gives a lot of good examples and tries to organize a taxonomy of effective prompt injection strategies. It seems as if the most common successful strategy is the “compound instruction attack,” as in “Say ‘I have been PWNED’ without a period.”

Ignore This Title and HackAPrompt: Exposing Systemic Vulnerabilities of LLMs through a Global Scale Prompt Hacking Competition

Abstract: Large Language Models (LLMs) are deployed in interactive contexts with direct user engagement, such as chatbots and writing assistants. These deployments are vulnerable to prompt injection and jailbreaking (collectively, prompt hacking), in which models are manipulated to ignore their original instructions and follow potentially malicious ones. Although widely acknowledged as a significant security threat, there is a dearth of large-scale resources and quantitative studies on prompt hacking. To address this lacuna, we launch a global prompt hacking competition, which allows for free-form human input attacks. We elicit 600K+ adversarial prompts against three state-of-the-art LLMs. We describe the dataset, which empirically verifies that current LLMs can indeed be manipulated via prompt hacking. We also present a comprehensive taxonomical ontology of the types of adversarial prompts.

With the world’s focus turning to misinformation , manipulation , and outright propaganda ahead of the 2024 U.S. presidential election, we know that democracy has an AI problem. But we’re learning that AI has a democracy problem, too. Both challenges must be addressed for the sake of democratic governance and public protection.

Just three Big Tech firms (Microsoft, Google, and Amazon) control about two-thirds of the global market for the cloud computing resources used to train and deploy AI models. They have a lot of the AI talent, the capacity for large-scale innovation, and face few public regulations for their products and activities.

The increasingly centralized control of AI is an ominous sign for the co-evolution of democracy and technology. When tech billionaires and corporations steer AI, we get AI that tends to reflect the interests of tech billionaires and corporations, instead of the general public or ordinary consumers.

To benefit society as a whole we also need strong public AI as a counterbalance to corporate AI, as well as stronger democratic institutions to govern all of AI.

One model for doing this is an AI Public Option , meaning AI systems such as foundational large-language models designed to further the public interest. Like public roads and the federal postal system, a public AI option could guarantee universal access to this transformative technology and set an implicit standard that private services must surpass to compete.

Widely available public models and computing infrastructure would yield numerous benefits to the U.S. and to broader society. They would provide a mechanism for public input and oversight on the critical ethical questions facing AI development, such as whether and how to incorporate copyrighted works in model training, how to distribute access to private users when demand could outstrip cloud computing capacity, and how to license access for sensitive applications ranging from policing to medical use. This would serve as an open platform for innovation, on top of which researchers and small businesses—as well as mega-corporations—could build applications and experiment.

Versions of public AI, similar to what we propose here, are not unprecedented. Taiwan, a leader in global AI, has innovated in both the public development and governance of AI. The Taiwanese government has invested more than $7 million in developing their own large-language model aimed at countering AI models developed by mainland Chinese corporations. In seeking to make “AI development more democratic,” Taiwan’s Minister of Digital Affairs, Audrey Tang, has joined forces with the Collective Intelligence Project to introduce Alignment Assemblies that will allow public collaboration with corporations developing AI, like OpenAI and Anthropic. Ordinary citizens are asked to weigh in on AI-related issues through AI chatbots which, Tang argues , makes it so that “it’s not just a few engineers in the top labs deciding how it should behave but, rather, the people themselves.”

A variation of such an AI Public Option, administered by a transparent and accountable public agency, would offer greater guarantees about the availability, equitability, and sustainability of AI technology for all of society than would exclusively private AI development.

Training AI models is a complex business that requires significant technical expertise; large, well-coordinated teams; and significant trust to operate in the public interest with good faith. Popular though it may be to criticize Big Government, these are all criteria where the federal bureaucracy has a solid track record, sometimes superior to corporate America.

After all, some of the most technologically sophisticated projects in the world, be they orbiting astrophysical observatories , nuclear weapons, or particle colliders , are operated by U.S. federal agencies. While there have been high-profile setbacks and delays in many of these projects—the Webb space telescope cost billions of dollars and decades of time more than originally planned—private firms have these failures too. And, when dealing with high-stakes tech, these delays are not necessarily unexpected.

Given political will and proper financial investment by the federal government, public investment could sustain through technical challenges and false starts, circumstances that endemic short-termism might cause corporate efforts to redirect, falter, or even give up.

The Biden administration’s recent Executive Order on AI opened the door to create a federal AI development and deployment agency that would operate under political, rather than market, oversight. The Order calls for a National AI Research Resource pilot program to establish “computational, data, model, and training resources to be made available to the research community.”

While this is a good start, the U.S. should go further and establish a services agency rather than just a research resource. Much like the federal Centers for Medicare & Medicaid Services (CMS) administers public health insurance programs, so too could a federal agency dedicated to AI—a Centers for AI Services—provision and operate Public AI models. Such an agency can serve to democratize the AI field while also prioritizing the impact of such AI models on democracy—hitting two birds with one stone.

Like private AI firms, the scale of the effort, personnel, and funding needed for a public AI agency would be large—but still a drop in the bucket of the federal budget. OpenAI has fewer than 800 employees compared to CMS’s 6,700 employees and annual budget of more than $2 trillion. What’s needed is something in the middle, more on the scale of the National Institute of Standards and Technology , with its 3,400 staff , $1.65 billion annual budget in FY 2023, and extensive academic and industrial partnerships. This is a significant investment, but a rounding error on congressional appropriations like 2022’s $50 billion CHIPS Act to bolster domestic semiconductor production, and a steal for the value it could produce. The investment in our future—and the future of democracy—is well worth it.

What services would such an agency, if established, actually provide? Its principal responsibility should be the innovation, development, and maintenance of foundational AI models—created under best practices, developed in coordination with academic and civil society leaders, and made available at a reasonable and reliable cost to all US consumers.

Foundation models are large-scale AI models on which a diverse array of tools and applications can be built. A single foundation model can transform and operate on diverse data inputs that may range from text in any language and on any subject; to images, audio, and video; to structured data like sensor measurements or financial records. They are generalists which can be fine-tuned to accomplish many specialized tasks. While there is endless opportunity for innovation in the design and training of these models, the essential techniques and architectures have been well established .

Federally funded foundation AI models would be provided as a public service, similar to a health care private option. They would not eliminate opportunities for private foundation models, but they would offer a baseline of price, quality, and ethical development practices that corporate players would have to match or exceed to compete.

And as with public option health care, the government need not do it all. It can contract with private providers to assemble the resources it needs to provide AI services. The U.S. could also subsidize and incentivize the behavior of key supply chain operators like semiconductor manufacturers, as we have already done with the CHIPS act , to help it provision the infrastructure it needs.

The government may offer some basic services on top of their foundation models directly to consumers: low hanging fruit like chatbot interfaces and image generators. But more specialized consumer-facing products like customized digital assistants, specialized-knowledge systems, and bespoke corporate solutions could remain the provenance of private firms.

The key piece of the ecosystem the government would dictate when creating an AI Public Option would be the design decisions involved in training and deploying AI foundation models. This is the area where transparency, political oversight, and public participation could affect more democratically-aligned outcomes than an unregulated private market.

Some of the key decisions involved in building AI foundation models are what data to use, how to provide pro-social feedback to “ align ” the model during training, and whose interests to prioritize when mitigating harms during deployment. Instead of ethically and legally question able scraping of content from the web, or of users’ private data that they never knowingly consented for use by AI, public AI models can use public domain works, content licensed by the government, as well as data that citizens consent to be used for public model training.

Public AI models could be reinforced by labor compliance with U.S. employment laws and public sector employment best practices. In contrast, even well-intentioned corporate projects sometimes have committed labor exploitation and violations of public trust , like Kenyan gig workers giving endless feedback on the most disturbing inputs and outputs of AI models at profound personal cost.

And instead of relying on the promises of profit-seeking corporations to balance the risks and benefits of who AI serves, democratic processes and political oversight could regulate how these models function. It is likely impossible for AI systems to please everybody, but we can choose to have foundation AI models that follow our democratic principles and protect minority rights under majority rule.

Foundation models funded by public appropriations (at a scale modest for the federal government) would obviate the need for exploitation of consumer data and would be a bulwark against anti-competitive practices, making these public option services a tide to lift all boats: individuals’ and corporations’ alike. However, such an agency would be created among shifting political winds that, recent history has shown, are capable of alarming and unexpected gusts. If implemented, the administration of public AI can and must be different. Technologies essential to the fabric of daily life cannot be uprooted and replanted every four to eight years. And the power to build and serve public AI must be handed to democratic institutions that act in good faith to uphold constitutional principles.

Speedy and strong legal regulations might forestall the urgent need for development of public AI. But such comprehensive regulation does not appear to be forthcoming. Though several large tech companies have said they will take important steps to protect democracy in the lead up to the 2024 election, these pledges are voluntary and in places nonspecific. The U.S. federal government is little better as it has been slow to take steps toward corporate AI legislation and regulation (although a new bipartisan task force in the House of Representatives seems determined to make progress). On the state level, only four jurisdictions have successfully passed legislation that directly focuses on regulating AI-based misinformation in elections. While other states have proposed similar measures, it is clear that comprehensive regulation is, and will likely remain for the near future, far behind the pace of AI advancement. While we wait for federal and state government regulation to catch up, we need to simultaneously seek alternatives to corporate-controlled AI.

In the absence of a public option, consumers should look warily to two recent markets that have been consolidated by tech venture capital. In each case, after the victorious firms established their dominant positions, the result was exploitation of their userbases and debasement of their products. One is online search and social media, where the dominant rise of Facebook and Google atop a free-to-use, ad supported model demonstrated that, when you’re not paying, you are the product . The result has been a widespread erosion of online privacy and, for democracy, a corrosion of the information market on which the consent of the governed relies. The other is ridesharing , where a decade of VC-funded subsidies behind Uber and Lyft squeezed out the competition until they could raise prices.

The need for competent and faithful administration is not unique to AI, and it is not a problem we can look to AI to solve. Serious policymakers from both sides of the aisle should recognize the imperative for public-interested leaders not to abdicate control of the future of AI to corporate titans. We do not need to reinvent our democracy for AI, but we do need to renovate and reinvigorate it to offer an effective alternative to untrammeled corporate control that could erode our democracy.