AI in Restaurants Goes Beyond the Phone: Palona AI Camera Integrates Orders and Operations into a Single Layer

The use of AI in restaurant technology has mostly been discussed through phone orders, chatbots, or customer service until now. Palona AI's new platform aims to take this approach to a broader operational layer.

The difference with the new approach is that it attempts to integrate phone, catering requests, sales signals, and operational data from within the physical store into the same system instead of automating a single customer channel.

AI Moves from Taking Orders to Monitoring Operations

Palona's product structure is divided into three main areas: Revenue Expansion, Revenue Intelligence, and Operations Intelligence/Excellence. According to the company's statement, the system captures phone calls, catering, and large order requests; analyzes their value and urgency; and tries to generate actions for managers.

What is more striking is the use of visuals.

According to SiliconANGLE, Palona's Operations Intelligence product can connect to existing restaurant security cameras. The system attempts to detect issues in customer wait times, cleanliness, or service flow from the footage and send alerts to managers.

This structure transforms restaurant AI from merely being a “system that talks to customers” into a layer that tries to understand physical operations.

What Does “Multimodal” Mean for Restaurants?

Multimodal AI refers to systems that can process text, audio, images, and various operational signals together instead of relying on a single data type.

The practical implications of this in a restaurant are significant.

A phone conversation can be an order signal. A camera image may indicate that wait times are increasing. A catering form can create a high-value potential sale. POS data can show whether the order was actually completed.

When these signals remain in disconnected systems, the manager has to track each one separately.

Palona claims to connect this data in a single loop of Capture → Understand → Act → Learn. However, independent, large-sample performance data regarding how reliable and scalable this approach is in a real restaurant environment is not yet publicly available.

Phone Calls Remain an Important Revenue Source

Palona's previous product was primarily built around conversational AI.

WIRED tested the “Jimmy the Surfer” AI system, which operates with Palona technology, for taking orders via phone and text at California-based Pizza My Heart in January 2025. The system suggests products to the customer, directs the order, and can chat in the brand's tone.

This example highlights a significant issue in restaurant technology.

Phone calls remain a revenue channel, especially for catering, group orders, and more complex requests. However, it is challenging for staff to answer every call during busy service times.

The value that AI can provide here may not be to completely eliminate the employee but to convert unanswered requests into sales.

The Company Reports a 20% Revenue Increase, but Independent Evidence is Limited

Palona claims that Father's Day revenue in the Cali BBQ application increased by more than 20% year-over-year and that the platform has become one of the business's significant sales channels. This result is based on the company's and customer manager's statements; it is not an independently audited study.

Therefore, it would not be accurate to generalize that "Palona increases restaurant revenue by 20%."

For a healthy assessment, at least the following questions need to be answered:

What would sales have been during the same period without AI? How much of the increase came from catering, and how much from better capturing existing demand? What is the system's cost? What is the contribution margin after food cost and labor for additional orders?

In restaurant technology, an increase in sales alone does not equate to ROI.

AI Using Cameras Brings New Risks

The most sensitive aspect of multimodal systems is image data.

Using a restaurant's existing security cameras for operational analysis can technically create new data collection and processing processes. It becomes crucial to understand how employees' and customers' images are processed, how long they are retained, who has access, and how local privacy regulations are complied with.

Palona's publicly available launch materials emphasize the operational benefits of the product; they do not explain the detailed legal framework of data protection and employee privacy practices in different countries.

Therefore, the purchasing decision for businesses should not end with the question, "What does AI see?"

Where is the data stored, how is the image transferred to the system, is employee notification required, and how is human oversight ensured? These questions should also be answered before the contract.

The AI Alert System Should Not Become a New Dashboard

Considering that restaurant managers already track numerous tools such as POS, delivery platforms, shift applications, inventory systems, and camera infrastructure, the real value of the new AI system is not to create another screen.

The value lies in the manager's ability to correctly identify the problem that requires intervention at the right moment.

Palona positions itself as a system that generates actions rather than “another dashboard.”

The operationally testable part of this claim is the quality of the alerts.

If the system generates too many false alerts, the manager may experience alarm fatigue. If it uses a very high threshold, it may miss real problems.

Therefore, in pilot projects, instead of measuring "how many alerts were sent?" it should be measured how many alerts actually turned into beneficial actions?

What is the Right Place for Automation from the Employee's Perspective?

The discussion of AI in restaurants is often reduced to job loss.

Palona's areas of use indicate another model: taking unanswered calls, classifying catering requests, or noticing operational signals that the manager might overlook.

When used correctly, these can reduce the repetitive burden on frontline employees.

However, if the technology turns into a surveillance tool that constantly monitors employee performance, the same system may weaken the employee experience.

The critical distinction for the business is this:

Does AI complete what the employee cannot do, or does it create an invisible oversight layer that constantly scores the employee?

This decision is more related to the business's usage policy than the technology provider.

What Should Be the Purchasing Criteria for Restaurants in Turkey?

Restaurant chains evaluating similar AI systems in Turkey need to define their problem before the product demo.

What percentage of phone orders are missed? From which channels do catering requests come? What operational problems are recurring in branches? Which data does the manager see too late?

If these questions are unanswered, a comprehensive AI platform can turn into an expensive technology layer.

In a healthier pilot model, the system can be tested in a few stores with limited KPIs: missed call rate, conversion, catering lead value, manager response time, and additional technology cost per order.

Technology should only be scaled if it creates meaningful improvement in these indicators.

Stafflink's View

In our opinion, the most significant change in Palona's announcement is the shift of restaurant AI from customer communication to the physical operations layer.

This potentially represents a significant leap. Because the toughest problems in restaurants arise not only digitally but also in physical processes such as queues, table turnover, cleanliness, employee coordination, and busy service.

However, the more data the system sees, the greater the responsibility of the business.

The right criterion for restaurants is not "how advanced is the AI?" but rather how accurately does it generate actions, does it genuinely reduce employee burden, can its ROI be measured, and can privacy risks be managed?