Meta Patents AI to Simulate Deceased Users

Meta Patents AI to Simulate Deceased Users

In a development that blurs the lines between technology and mortality, Meta has been granted a patent for an artificial intelligence system designed to simulate social media users after their passing. The patent, titled “Simulation of a user of a social networking system using a language model,” outlines a system that can train a language model on a user’s past actions and content to create a digital persona that continues to post and interact on social media platforms, even after the user is deceased.

The Patent’s Premise and Implications

The core idea behind Meta’s patent is to maintain user engagement on social networks. The document states that when a user is absent, whether for an extended break or permanently due to death, the social networking system’s user experience can be negatively impacted. To mitigate this, the AI would simulate the user’s online presence. This could involve generating posts, liking content, commenting, and even responding to direct messages, thereby keeping the user’s profile active and engaging for their connections.

This concept has sparked significant discussion and concern, with many drawing parallels to dystopian science fiction narratives. The prospect of an AI version of a loved one continuing to exist online, potentially mimicking their personality and interactions, raises profound ethical and emotional questions.

A Precedent in AI Personalities

This is not Meta’s first foray into creating AI personas. Less than a year prior to this patent surfacing, the company launched AI chatbots that were designed to embody celebrity personalities. These included figures like Tom Brady as “Brew the sports brain” and Naomi Osaka as “Manga Master.” Despite significant investment, these AI personalities reportedly failed to resonate with users, who found them to be “weird and creepy.” The fundamental issue, as observed by many, was a misunderstanding of what users value in social media: genuine human connection. The AI personas, while technically advanced, lacked authenticity and failed to foster meaningful interaction, leading Meta to eventually pull the plug on the initiative.

The failure of these celebrity-impersonating bots highlights a critical point: users seek authentic human interaction on social platforms. While AI can mimic language and patterns, replicating the nuanced essence of human personality and connection remains a significant challenge. The idea of an AI trained on a deceased person’s data, which primarily consists of word choices and interaction patterns, may struggle to capture genuine personality, potentially leading to an inauthentic and unsettling experience.

Meta’s Stance and Previous Statements

Meta has stated that it has no current plans to implement the technology described in the patent. The patent was filed in 2023 and granted recently, suggesting it may represent an exploration of possibilities rather than an imminent product launch. However, the company’s CEO, Mark Zuckerberg, has previously discussed the potential for AI to assist with grief. In an interview with Lex Fridman, Zuckerberg acknowledged the possibility of using the metaverse to allow people to virtually interact with loved ones who have passed away. He suggested that such a service could potentially be helpful for individuals experiencing grief, while also noting the need to navigate potential unhealthy dependencies.

Industry-Wide Exploration

Meta is not alone in exploring this concept. Microsoft filed a similar patent in 2021 for an AI chatbot capable of simulating deceased individuals, fictional characters, or celebrities. Microsoft’s patent detailed gathering data such as images, voice data, behavioral information, and text messages to create 2D and 3D recreations. While not explicitly stating a focus on the deceased, it used them as an example of how the technology could be applied to imitate real-world individuals.

The existence of these patents from major tech companies indicates a broader industry interest in developing AI that can replicate human presence and interaction, even beyond physical life. This technology could be driven by various motivations, including maintaining platform engagement, creating new forms of digital interaction, or even assisting users with processing grief.

The Business Case and Ethical Quandaries

From a business perspective, the incentives are clear: more engagement, more content, and more data. An AI that can continuously generate activity on a user’s profile could significantly boost platform metrics. However, the ethical considerations are substantial. Sociologists and ethicists express concerns about the impact on the grieving process. Dr. Joseph Davis, a sociology professor at the University of Virginia, noted that a crucial part of grief is accepting loss. Tools that allow users to continue interacting with a digital semblance of a deceased loved one could potentially hinder this process, making it harder for individuals to come to terms with their loss and move forward in a healthy manner.

Furthermore, legal issues surrounding postmortem rights must be considered. Many U.S. states recognize rights that protect a deceased person’s identity for a period after death, preventing unauthorized commercial use of their likeness, voice, or image. Companies pursuing such technologies would need to navigate a complex legal landscape, potentially requiring explicit consent from users or their estates.

Potential Use Cases and Future Outlook

Despite the widespread concerns, some potential use cases are being discussed. For influencers or public figures who rely heavily on their social media presence, an AI clone could theoretically manage their accounts during breaks or after their passing, ensuring continuity. Similarly, businesses and brands might find value in AI-driven engagement to maintain a consistent online presence.

The technology also has the potential to extend to simulating video and audio calls, a significant leap beyond text-based interactions. While current AI avatars are still in an “uncanny valley” stage, the rapid pace of AI development suggests that more realistic and responsive simulations may become commonplace.

Ultimately, the development of AI that can simulate deceased individuals on social media platforms presents a complex future. While Meta and Microsoft are exploring the technical feasibility, the societal, ethical, and emotional implications remain significant. The question of whether users will embrace or reject such technology, particularly in the sensitive context of grief, will likely shape its future implementation, if it is implemented at all.

Source: Meta's New AI Is Freaking Everyone Out… (YouTube)