Imagine finding out something went wrong after placing an order for a birthday cake from a local bakery.
The store is busy when you try to message the owner. The proprietor serves clients, takes calls, and attempts to complete the day’s orders.
What if artificial intelligence could review your order and generate a useful response in just seconds?
Meta Platforms (META) wants to make the technology behind those everyday tasks dramatically cheaper.
The Facebook parent introduced Muse Ember on Nov. 9, pricing its new AI model at just 10 cents per million input tokens, the small pieces of information it processes. For every million output tokens, the cost of generating answers is an additional 50 cents.
This might result in less paperwork, more client time, and less expensive customer service for small firms.
Although it may result in less monotonous tasks for employees, it raises concerns about job security. It also poses a bigger concern for Meta shareholders: Can the firm transform low-cost AI into a successful venture?
Meta’s official announcement confirmed the pricing and availability of Muse Ember.
There is a crucial catch. Developers, not regular consumers, are the target audience for this technology. Businesses must develop or acquire applications that use it.
Meta’s 10-cent AI could make small businesses more efficient
Imagine an online business that responds to thousands of inquiries from customers each month. Certain clients request refunds. Others want assistance locating items or shipping updates.
AI may arrange such inquiries, gather pertinent data, and prepare answers for staff members to examine.
It may be surprisingly inexpensive. Let’s say a company receives 10,000 requests with 1,000 input tokens and responds with 250 tokens.
The model fees would come to around $2.25 at Muse Ember’s regular prices. That amount does not include software development, staff management, and other operational costs.
Meta’s official pricing documentation also shows that requests exceeding 200,000 input tokens cost more.
However, reduced processing costs can encourage smaller businesses to use technology that they previously thought was prohibitively expensive.
Related: Democrats prepare to go after Apple, Meta and Elon Musk companies
After completion, a contractor might arrange the bills. A retailer might handle customer concerns. A dental practice might categorize appointment requests.
These are not services that Meta has explicitly declared; rather, they represent possible applications.
Meta’s cheaper AI could change jobs and customer service
A number of businesses have already tried Muse Ember. Box, a cloud software company, found positive outcomes from applying the approach to evaluate employment criteria and analyze contracts.
According to Cloudflare, Muse Ember outperformed a similar model by up to 30% in several tests. Meta’s announcement included those early findings.
More Meta:
- Meta’s $100 billion stock sale shrinks your slice of the company
- Bank of America revamps Meta stock price target once again
- Meta stock falls despite Wall Street analysts raising price targets
For customers, faster responses could mean less waiting for basic assistance. However, speed does not equate to quality.
An AI system might misinterpret a refund request or provide false information. Safeguards and human supervision would still be necessary for businesses.
Another issue with technology handling sensitive papers, financial information, or personal data is privacy.
Employees deal with a distinct kind of uncertainty. They might spend more time resolving complex consumer issues if AI takes care of simple administrative tasks. But companies may also leverage such efficiencies to cut employment costs.
In addition, lower operational expenses may not always translate into lower client prices. Instead of lowering prices, businesses might keep the savings.
Meta’s latest AI breakthrough has a surprising price tag.
Meta’s $145 billion AI investment faces a profitability test
Cheaper AI presents a costly paradox for Meta.
The business recorded $64.6 billion in sales for the third quarter, up 26% from the same period last year. Yet capital expenditures reached $34.91 billion.
Meta’s October earnings release also projected full-year capital spending between $135 billion and $145 billion. There is pressure to provide significant financial returns given that massive expenditure.
Muse Ember might draw businesses looking for reasonably priced AI services while lowering the cost of operating some of Meta’s current technologies. But reduced costs do not equate to large profits.
Meta has to create enough revenue to pay for infrastructure and development costs.
Investors should monitor recurring revenue, client uptake, and whether AI investments increase profitability.
The choice is more urgent for small company owners: Does automation save enough time and money to warrant its setup?
The challenge for workers is whether AI simplifies their work or takes over some duties.
Additionally, consumers could discover the solution during their subsequent regular encounter with a company.
The real test of Meta’s 10-cent AI isn’t how cheaply it processes information. It’s whether businesses can turn those savings into better service without sacrificing reliability.
Related: Meta’s $100 billion stock sale shrinks your slice of the company
