Former X product executive Nikita Bier has raised questions about how Meta’s new Muse artificial intelligence agent could affect Facebook Marketplace and the future of online buying and selling.
Bier, who previously served as a product executive at X, has suggested that the growing use of AI agents could fundamentally change the way transactions take place on platforms such as Facebook Marketplace. His comments focus on a possible situation in which AI agents begin negotiating with sellers instead of human buyers.
The comments come shortly after Meta introduced Muse as a personal AI agent designed to perform tasks on behalf of users. Meta says Muse can take action across applications, use a browser, complete forms and negotiate on behalf of users.
The technology could eventually have implications for online marketplaces, where buyers and sellers currently communicate directly with one another. If AI agents become responsible for finding products, comparing prices and negotiating deals, the traditional human-to-human interaction that has characterised many Marketplace transactions could change significantly.
Bier has warned that the widespread use of automated buyers could create challenges for sellers. One concern is that large numbers of AI agents could automatically contact sellers and make offers without the normal human considerations that influence negotiations.
For example, an AI agent acting on behalf of a buyer could be programmed to search for the lowest available price and negotiate aggressively. If millions of such agents were active simultaneously, sellers could receive a large number of automated offers, potentially making the buying and selling process different from the current experience.
Bier has used strong language to describe the possible impact, suggesting that Meta’s AI could potentially “kill” Facebook Marketplace. However, this is his prediction about a possible future effect rather than an established outcome.
Meta itself has not said that Muse is intended to replace human interactions on Facebook Marketplace. Instead, the company has presented Muse as a general-purpose personal AI agent that can help users perform everyday tasks.
According to Meta, Muse can work on a user’s behalf across different applications. It can open a browser, fill out forms and coordinate tasks. The company also says Muse can negotiate on behalf of a user, which is one of the features that has attracted attention in relation to online commerce.
Meta introduced Muse on September 8, describing it as a personal AI agent designed to help people accomplish tasks rather than simply answer questions.
The company says Muse runs through a dedicated secure virtual machine known as Muse Secure VM. Meta says this environment is designed to isolate the agent and protect user data and credentials.
Muse can be instructed to carry out tasks such as sending emails, booking travel and making purchases. Meta says the system can continue working after a user closes the application and return when something changes or when approval is required.
The company's description also includes shopping-related functions. Meta says Muse can help users find deals and negotiate on their behalf. The agent can also use payment systems to complete eligible purchases after appropriate user approval.
These capabilities create a new possibility for online marketplaces. Instead of a person manually searching through listings, contacting sellers and negotiating prices, an AI agent could potentially perform some or all of these steps.
Facebook Marketplace is already using artificial intelligence in several parts of its platform. Meta introduced AI-powered features that can help sellers create listings, suggest prices and respond to buyer inquiries.
Meta says sellers can upload photographs of an item and use AI to create a draft listing with details and pricing suggestions based on similar items in the area. The company has also introduced AI-generated replies that can respond to common buyer questions about availability, location and price.
These existing tools show that AI is already being incorporated into the Marketplace experience. However, the new generation of personal AI agents could go further by allowing the buyer's software to actively search for products and negotiate with sellers.
This distinction is important because an AI assistant that helps a seller write a response is different from an autonomous agent that acts as a buyer.
If buyers increasingly use agents to negotiate automatically, sellers may have to adapt to conversations that are partly or entirely automated. This could change how sellers determine prices, respond to offers and decide whether a buyer is genuinely interested.
There could also be benefits. AI agents could make it easier for buyers to compare prices across listings and identify products that meet specific requirements. Sellers could potentially reach more buyers without spending as much time responding to repetitive questions.
Automation could also reduce the amount of routine communication required from both sides. A buyer could specify a budget and preferred features, while an AI agent could search available listings and identify potential matches.
However, concerns could emerge if automated systems generate large numbers of low offers. Sellers could face increased pressure to respond to automated negotiations, particularly if agents are designed to prioritise the lowest possible price.
The issue also raises questions about transparency. A seller may not always know whether they are negotiating with a person or an AI system. Similarly, buyers may not know whether a seller's response has been written by a person or generated automatically.
Another issue is the possibility of automated competition between AI agents. If multiple agents are independently searching for products and negotiating prices, marketplaces could eventually become environments where software systems conduct a significant portion of transactions.
That scenario remains hypothetical, and there is no evidence that Facebook Marketplace has already reached such a stage.
Meta's current Muse rollout is also limited by the company's stated availability and controls. Meta says Muse is initially rolling out in the United States on iOS, Android and muse.ai, with availability on AI glasses planned for the future.
The company has also built approval mechanisms into Muse. Meta says users can decide which applications the agent can access and how much permission it receives. The system can ask for approval before sensitive actions such as sending an email or making a purchase.
These controls are relevant to the Marketplace discussion because they mean users are not necessarily handing complete control of their transactions to an AI agent without oversight.
At the same time, the broader development of AI agents could have implications for digital commerce beyond Facebook Marketplace. Other marketplaces and shopping platforms may also need to adapt if consumers increasingly use autonomous software to search for products and negotiate purchases.
The future relationship between AI agents and online marketplaces is therefore still developing. Companies may need to establish new rules governing automated purchasing, seller communication, price negotiations and transaction security.
For Facebook Marketplace, the potential change could be particularly significant because the platform is built around interactions between individual buyers and sellers. Much of its appeal comes from users discovering items, asking questions, negotiating prices and arranging transactions.
If AI agents perform a growing share of those activities, the character of the platform could change.
Bier’s comments highlight this possibility, but they should be understood as an observation and prediction rather than a confirmed assessment of Meta Marketplace's future.
Meta's own position is that AI agents can make everyday tasks easier and help people accomplish goals with less effort. The company sees Muse as a step toward personal AI systems that can perform practical tasks on behalf of users.
The debate over Facebook Marketplace illustrates one of the broader questions surrounding this technology. AI agents may save users time, but widespread automation could also change the way people communicate and negotiate online.
Whether that results in a more efficient marketplace or creates new challenges for sellers will depend on how users adopt the technology, how marketplaces regulate automated activity and how AI companies design their agents.
For now, Facebook Marketplace continues to operate as a platform connecting buyers and sellers, while Meta is adding more AI-powered tools to improve listing creation and communication.
The introduction of Muse represents a potentially larger step toward automated commerce because Meta says the agent can take actions and negotiate on behalf of users.
Bier’s warning therefore focuses on a possible future in which AI agents become common participants in online marketplaces. The idea that this could “kill” Facebook Marketplace remains his prediction and is not a confirmed development.
As AI agents become more capable, the interaction between people and online marketplaces is likely to evolve. Buyers may increasingly rely on software to search, compare and negotiate, while sellers may need new tools to manage automated inquiries.
The impact on Facebook Marketplace will ultimately depend on how quickly consumers adopt these agents and how Meta integrates them with its commerce platforms.