Accor Deploys ALL Concierge, Its Global Conversational AI Assistant


Accor déploie « ALL Concierge », son assistant IA conversationnel à l’échelle mondiale - Depositphotos.com @MuhammadAlimaki

Aer Lingus


Accor moves to a new stage in its artificial intelligence strategy with the deployment of ALL Concierge.

After a pilot phase that generated more than a million conversations, the hotel group is extending this conversational assistant to its main channels, with the goal of supporting customers before, during and after their stay.

Accessible 24/7, ALL Concierge guides the traveler from inspiration to managing their reservation.

The tool can notably propose destinations and hotels according to expressed preferences, compare establishments, provide practical information about services, or guide directly to the booking.

The assistant can also answer questions related to an existing reservation and recognize ALL Accor loyalty program members in order to tailor its responses to their profile and past stays.


Accor: a pilot phase of over a million conversations

Launched in the pilot phase in July 2025, ALL Concierge drew on the analysis of more than one million customer conversations to refine its features.

Accor notes that travelers’ expectations extended beyond simple customer service, with a desire for inspiration, comparison and guidance throughout the journey.

The service is now available on ALL.com in 11 languages, as well as on the ALL Accor app and messaging platforms such as WhatsApp and iMessage. Accor plans to gradually extend its availability to all languages covered by the group by the end of the year.

The AI is however not designed as a systematic substitute for the teams. When a request requires human intervention, the customer can be transferred to an Accor colleague.

The group also notes that certain situations identified as sensitive may trigger this handover.


Des fonctionnalités appelées à s’élargir

The scope of ALL Concierge still needs to evolve in the coming months. Accor plans notably more advanced comparison features, integrating AI-synthesized customer reviews, as well as recommendations for activities and local events proposed with its partners.

The service should also progressively enable ALL Accor members to optimize the use of their points and benefits, to modify or cancel a booking directly from the conversational interface, and then access their invoices and loyalty information after their stay.

In the long term, Accor also intends to enable a voice interaction with ALL Concierge. Meeting and event spaces are to be integrated into the solution at a later stage.


The “Butler” for hotel teams

In parallel, Accor is developing “Butler,” an AI assistant designed for hotel staff. It is intended to give them quick access to information, advice and recommendations, but also to summon the right experts when intervention is necessary.

This initiative fits into Accor’s broader AI strategy. According to the group, 82% of its headquarters staff currently use AI tools in their daily activities. Accor also says it aims to automate certain internal tasks to free up time for interactions with guests.

At the conference, Alix Boulnois, Chief Commercial, Digital & Tech Officer of Accor, summarized this approach around the idea of “automating the ordinary” to leave more room for the human relationship in hospitality.

Finally, regarding data, Accor says it relies on several internal agents and different AI models, while ensuring GDPR compliance. The group notes that data are used in aggregated form to identify certain trends and to personalize interactions with customers.


Amelia Brille Publié par Amelia Brille Rédactrice TourMaG.com
Voir tous les articles d’Amélia Brille

  • picto Linkedin
  • picto email
Ajoutez TourMaG à votre flux Google Actualités Google Actualités icône  

Mirai Integrates Multi-Room Booking into Its AI Assistant Sarai


Mirai fait évoluer Sarai pour gérer les demandes de plusieurs chambres - Depositphotos.com, sdecoret

CroisiEurope


Mirai upgrades its AI assistant and its conversational reservations engine Sarai.

The solution can now handle requests for reservations of multiple rooms within a single conversation, without forms or intermediate steps.

A client can thus directly formulate a request such as: “I need two rooms for a family with children”. Sarai is then able to identify the need, manage the booking, and continue the exchange in natural language.

This feature is available across all channels on which Sarai operates, including the webchat, WhatsApp, email, Instagram and telephone.

It is also accessible via ChatGPT, thanks to Apps, and Claude, via MCP.

Mirai states that it is included at no additional cost for customers of its reservations engine.


Multi-room bookings account for 18% of revenue

For hotels, requests involving multiple rooms represent a significant commercial challenge.

According to Mirai’s data up to July 2026, bookings for two rooms or more account for 9.07% of the total reservations made via the web engine.

They nevertheless generate 18.45% of the nights and 18.44% of the revenue linked to rooms. Their value weight is therefore almost twice their volume weight.

The demand is also present on conversational channels. During the last 30 days, 7.03% of availability requests processed by Sarai involved multiple rooms.


Group requests turned into qualified leads

The new feature also accounts for requests that exceed the number of rooms that can be booked online directly.

In this case, Sarai explains to the client the online booking limits, collects their contact details as well as the details of their stay, and then generates a qualified lead for the hotel.

The request is then visible in real time in Lobby, the AI infrastructure dashboard of Mirai, and triggers an email notification to the establishment.

For Mirai, this approach allows transforming a request that could have been abandoned with a traditional booking engine into a business opportunity at the moment the client expresses their purchase intent.

“Handling multi-room bookings was essential in conversational reservations. Today, customers express their need with all their details and their complexity in natural language, and expect the hotel to know how to respond“, explains Pablo Delgado, CEO America of Mirai, in a press release.

According to him, available data show that hotels capable of handling these complex requests are able to attract higher-value clientele.


Amelia Brille Published by Amelia Brille TourMaG.com editor
See all articles by Amelia Brille

  • picto Linkedin
  • picto email
Add TourMaG to your Google News feed Google Actualités icône