Selected recommendation
The still centre
Thousands of cabins. Thousands of different days.
Guest, moment and ship, weighed in one place.
One recommendation. Or none at all.
The mind behind the moment.
AIfred is Experiential AI for cruise and hospitality: it judges every recommendation by the guest’s experience, not by the click.
AIfred brings together who a guest is, how they are right now and what is actually available on board. It chooses one recommendation, delivers it through the touchpoints you already run, and ties every result back to the call that earned it.
The system at a glance
From guest context to one relevant action.
AIfred sits between the information you already have and the touchpoints your guests and teams already use. It combines context, weighs relevance and selects one recommendation.
01
Guest
Who is this person?
Cabin 7115. Third sailing, a regular in the spa.
02
Moment
What matters right now?
A short, restless night. A free sea-day morning.
03
Ship
What is actually possible?
The spa has a place at 10:30.
decides
AIfred
Weighs all three and selects one recommendation, or none at all.
One quiet line on her app, 08:40.
04
Outcome
Delivered, observed, learned from.
Spa booked, tied to the call.
Reactive service waits to be asked. Proactive systems fire rules at everyone. AIfred is anticipatory: it acts on how the guest is right now.
Context, not just a profile
A profile says who the guest is. The moment says what matters now.
Your systems already know the booking. AIfred adds what is true right now, so a recommendation fits the day the guest is actually having.
What you already know
- Preferences spa, quiet mornings
- Bookings nothing today
- Voyage history third sailing
- Party and occasion travelling solo
What matters now
- How the guest is: sleep, recovery short, restless night
- Presence in the cabin in since 07:30
- Time and place in the voyage sea day, 08:40
- Live availability on board spa free at 10:30
- Operational context pool deck busy from 11:00
AIfred combines both.
The decision
Many possibilities. One decision.
AIfred weighs every eligible option against the guest, the moment and what is actually free, and selects the one that fits best.
Eligible this morning
Spa treatmentWellness · 10:30
SelectedThermal suiteWellness · 16:30
ConsideredChef’s tableDining · 19:30
Later todayShore excursionExperience · tomorrow
ConsideredPool bar happy hourBar · 11:00
Not a fit
Four questions
01
Guest fitDoes it suit this guest?
02
MomentDoes it suit how they are right now?
03
AvailabilityCan it actually be booked?
04
Operational senseDoes it make sense for the ship today?
One recommendation
If nothing is relevant enough, AIfred recommends nothing.
Relevance comes before volume. A quiet screen is better than a wrong offer, for the guest and for every offer that follows.
Illustrative. Offers, times and cabin are examples.
From decision to outcome
Every result, tied back to the call that earned it.
AIfred makes the decision. Your existing guest and crew touchpoints stay the interface. Every response flows back.
Decision
AIfred
Selects one recommendation.
Spa at 10:30, for Cabin 7115.
Touchpoint
Your screens
Guest app, cabin TV or crew device.
Her guest app, 08:40.
Response
Guest action
Accepted, declined or ignored.
Accepted at 08:52.
Outcome
Attributed result
Booked, tied to its recommendation, measured.
Spa booked, revenue tied to this call.
Feedback returns to AIfred. Each outcome stays linked to the recommendation that created it, and sharpens the next one. At 15:10 her usual massage is full, so the next call is the thermal suite at 16:30.
The proof
Every call carries its reasoning.
Which reasons, and how they are weighed, we walk through with you in a briefing.
Sailing
Sea day 4 / 7
Queue
Cabin 7115
Cabin 9031
Suite 12A
7115
Cabin 7115 · Deck 7
Sea day · her last evening at sea
Offer two seats at the chef’s table tonight at 19:30, early enough to say yes.
Confidence
0.92
Guest staterelaxed
Contextlast evening, no dinner booked
Availabilitychef’s table, 19:30
The full decision trail is shared in a briefing, under NDA. Request one.
token ••••
Illustrative. Cabin, sailing and values are examples.
The components
One AIfred. Three parts.
Sense picks up the raw signals. Mate turns them into a signal AIfred can use. AIfred brings it all together and makes the call. Three parts, one system.
01 Part of AIfred
sealevel.Sense
The sensesPicks up the raw signals in the cabin: presence, sleep and recovery. The things a booking system never sees.
02 Part of AIfred
sealevel.Mate
The home on boardTurns the raw signals into something AIfred can use, on the ship itself, and hands it over.
03 The centre
sealevel.AIfred
The mindBrings everything together, makes the right call, or no call at all, and learns from what follows.
Cabin 7115 · 08:40 Sense picks up a short, restless night. Mate turns it into a clear signal. AIfred suggests the spa at 10:30.
Integration
Something new on board. Nothing to rip out.
AIfred is not one more layer in your stack. It is a new kind of intelligence on board, one that understands the moment, decides and learns from every outcome. It connects to the systems you already run and speaks through the screens your guests already use.
in
Your systems
- Guest profiles and bookings
- Availability and capacity
- Operational data
runs onboard on sealevel.Mate
AIfred
- Context processing
- Decision logic
- Recommendation and attribution
out
Your touchpoints
- Guest app
- Cabin TV
- Crew devices
- Other digital channels
Trust and control
Built for real ships.
And for the procurement review that comes with them.
Data stays at sea level
Guest data is processed within your environment and your data-residency rules. We hold nothing we do not need.
Sensitive by design
Sleep and recovery signals are treated as sensitive data. A data-protection impact assessment is part of every pilot scope.
Auditable by design
Each recommendation keeps its token and trail, so any call can be reconstructed and defended later.
Your data, made easy to read
Through the pilot you keep full access to the data you need, and we sit with your team to turn it into clear insight and reporting.
Every decision. Shaped together.
Share what matters to you: a guest moment, a challenge or an ambition. Together, we’ll trace how AIfred would read it, which call it would make and how the result ties back to what it earned.