Manifesto

Preface

User interfaces have evolved over time, from command lines used by technical operators to graphical interfaces that made insurance software usable at scale. That shift unlocked efficiency, but it also locked companies into portals, menus, and rigid workflows.

Customer behavior has changed since then. When something breaks, when a quote stalls, when a claim feels unclear, people do not want another portal or a long call queue. They want a simple interaction that moves things forward.

Insurance is now feeling the gap between how work is structured internally and how customers expect to engage.

Thesis

We believe the next shift in insurance operations will happen in messaging.

More of the work that currently happens across calls, emails, and portals will move into simple text conversations that can actually complete tasks. Not notifications. Not reminders. Real work.

SMS is the most reliable place to do this. It is where urgency shows up, where customers respond, and where attention already exists when something matters.

General Magic is building infrastructure to make messaging a dependable operating channel for insurers.

The current state

Across the industry, customers are already trying to interact this way. They reply to reminder texts with questions. They send documents over text. They ask for changes outside of office hours.

Most insurance systems are not built to handle that. SMS is treated as a one way notification pipe, while the actual work still requires a call, a portal login, or manual follow up from a team member.

The result is predictable. Quotes go cold. Claims stall. Documents are chased. Call centers absorb avoidable volume. Teams spend time coordinating instead of resolving.

Why this has not worked before

Messaging sounds easy until it has to touch real insurance workflows.

The moment you need identity checks, document collection, policy details, billing changes, endorsements, or exceptions, everything breaks unless the conversation is deeply connected to systems of record.

On top of that, this is not something a team can ship once and forget. Reliability matters. Accuracy matters. One wrong answer at the wrong moment destroys trust.

Most insurers do not want to build and maintain this layer themselves. It requires AI expertise, ongoing iteration, operational discipline, and deep integration work.

The AI messaging layer for insurance

We believe there is a new category here.

Insurers will operate customer facing AI agents that live in messaging and are directly connected to their internal systems. On the company side, the product looks like an operational workspace where each conversation represents a workflow in motion. On the customer side, it becomes a real alternative to calling, waiting, and bouncing between departments.

Some workflows are handled dynamically in simple cases. More complex flows are designed ahead of time and triggered by customer messages or internal events like quotes, FNOL, or renewals.

Why insurance first

Insurance is one of the most procedural industries in existence.

Work is stateful, multi step, and full of edge cases. Human teams act as glue between systems, not because judgment is always required, but because coordination is.

You see it everywhere. Quote to bind. FNOL to adjust to approve to pay. Endorsements. Renewals. Document chasing. Follow ups.

At the same time, insurance already has strong systems of record. Policy systems. Claims systems. Billing. CRMs and agency management systems. The logic exists. The breakdown happens at the point of customer interaction.

Where General Magic fits

We offer a product called Cell, and we apply it at the moments where insurance workflows most often stall.

Cell powers SMS agents that insurers deploy after a quote, at first notice of loss, during servicing, and through renewal. The agent answers questions, collects documents, runs verification, handles KYC flows, and moves workflows forward without another call.

On the insurer side, this is not about replacing teams. It is about reducing manual follow up, smoothing handoffs, and making outcomes more consistent.

What AI adds on top of basic SMS

Basic SMS sends reminders. AI turns reminders into completed actions.

Instead of telling someone to remember an appointment, the agent can confirm or reschedule in chat, handle conflicts immediately, and answer the small questions that cause no shows. That removes ambiguity and friction, which is where most drop off happens.

AI also enables personalization that actually matters. Not surface level fields, but adjustments based on behavior, constraints, language, timing, and intent. That builds trust and improves follow through.

Forms become conversations. Structured data is collected naturally and mapped directly into systems, which means fewer portals, fewer logins, and fewer abandoned flows.

AI provides just in time guidance for procedural work. It breaks tasks into small actions, detects confusion, and explains next steps clearly. That is what reduces delays and “I will do it later” behavior.

When things get complex, escalation works. The agent detects risk, routes to a human with full context, and resumes the workflow after handoff. That is how messaging becomes dependable in real operations.

What needs to be true to win

For messaging to replace calls, it has to be faster than doing the work manually. It has to be reliable enough that teams trust it in high stakes moments. It has to feel native to the insurer’s brand and flexible to individual customer needs.

Most importantly, it has to be connected to systems of record. Messaging without action is just a nicer FAQ.

Time to value also matters. Insurers should see impact in the first conversations, not after months of setup.

Looking ahead

We expect General Magic to take on more of the customer interaction layer over time, not by owning more screens, but by owning the moments where customers are stuck, confused, or at risk of dropping out.

Insurance is the starting point because the need is clear and the impact is immediate. The same approach applies anywhere follow through breaks down and coordination becomes costly.

The opportunity is not just operational efficiency. It is a better experience for customers and calmer, more focused teams for insurers.

Parting words

We believe messaging will become a core operating channel for insurance, not as a support tool, but as a place where real work gets done.

When implemented correctly, AI driven messaging reduces friction, recovers lost workflows, and lowers operational load, while making insurance feel more responsive and human at the moments that matter.

A & JPreface

User interfaces have evolved over time, from command lines used by technical operators to graphical interfaces that made insurance software usable at scale. That shift unlocked efficiency, but it also locked companies into portals, menus, and rigid workflows.

Customer behavior has changed since then. When something breaks, when a quote stalls, when a claim feels unclear, people do not want another portal or a long call queue. They want a simple interaction that moves things forward.

Insurance is now feeling the gap between how work is structured internally and how customers expect to engage.

Thesis

We believe the next shift in insurance operations will happen in messaging.

More of the work that currently happens across calls, emails, and portals will move into simple text conversations that can actually complete tasks. Not notifications. Not reminders. Real work.

SMS is the most reliable place to do this. It is where urgency shows up, where customers respond, and where attention already exists when something matters.

General Magic is building infrastructure to make messaging a dependable operating channel for insurers.

The current state

Across the industry, customers are already trying to interact this way. They reply to reminder texts with questions. They send documents over text. They ask for changes outside of office hours.

Most insurance systems are not built to handle that. SMS is treated as a one way notification pipe, while the actual work still requires a call, a portal login, or manual follow up from a team member.

The result is predictable. Quotes go cold. Claims stall. Documents are chased. Call centers absorb avoidable volume. Teams spend time coordinating instead of resolving.

Why this has not worked before

Messaging sounds easy until it has to touch real insurance workflows.

The moment you need identity checks, document collection, policy details, billing changes, endorsements, or exceptions, everything breaks unless the conversation is deeply connected to systems of record.

On top of that, this is not something a team can ship once and forget. Reliability matters. Accuracy matters. One wrong answer at the wrong moment destroys trust.

Most insurers do not want to build and maintain this layer themselves. It requires AI expertise, ongoing iteration, operational discipline, and deep integration work.

The AI messaging layer for insurance

We believe there is a new category here.

Insurers will operate customer facing AI agents that live in messaging and are directly connected to their internal systems. On the company side, the product looks like an operational workspace where each conversation represents a workflow in motion. On the customer side, it becomes a real alternative to calling, waiting, and bouncing between departments.

Some workflows are handled dynamically in simple cases. More complex flows are designed ahead of time and triggered by customer messages or internal events like quotes, FNOL, or renewals.

Why insurance first

Insurance is one of the most procedural industries in existence.

Work is stateful, multi step, and full of edge cases. Human teams act as glue between systems, not because judgment is always required, but because coordination is.

You see it everywhere. Quote to bind. FNOL to adjust to approve to pay. Endorsements. Renewals. Document chasing. Follow ups.

At the same time, insurance already has strong systems of record. Policy systems. Claims systems. Billing. CRMs and agency management systems. The logic exists. The breakdown happens at the point of customer interaction.

Where General Magic fits

We offer a product called Cell, and we apply it at the moments where insurance workflows most often stall.

Cell powers SMS agents that insurers deploy after a quote, at first notice of loss, during servicing, and through renewal. The agent answers questions, collects documents, runs verification, handles KYC flows, and moves workflows forward without another call.

On the insurer side, this is not about replacing teams. It is about reducing manual follow up, smoothing handoffs, and making outcomes more consistent.

What AI adds on top of basic SMS

Basic SMS sends reminders. AI turns reminders into completed actions.

Instead of telling someone to remember an appointment, the agent can confirm or reschedule in chat, handle conflicts immediately, and answer the small questions that cause no shows. That removes ambiguity and friction, which is where most drop off happens.

AI also enables personalization that actually matters. Not surface level fields, but adjustments based on behavior, constraints, language, timing, and intent. That builds trust and improves follow through.

Forms become conversations. Structured data is collected naturally and mapped directly into systems, which means fewer portals, fewer logins, and fewer abandoned flows.

AI provides just in time guidance for procedural work. It breaks tasks into small actions, detects confusion, and explains next steps clearly. That is what reduces delays and “I will do it later” behavior.

When things get complex, escalation works. The agent detects risk, routes to a human with full context, and resumes the workflow after handoff. That is how messaging becomes dependable in real operations.

What needs to be true to win

For messaging to replace calls, it has to be faster than doing the work manually. It has to be reliable enough that teams trust it in high stakes moments. It has to feel native to the insurer’s brand and flexible to individual customer needs.

Most importantly, it has to be connected to systems of record. Messaging without action is just a nicer FAQ.

Time to value also matters. Insurers should see impact in the first conversations, not after months of setup.

Looking ahead

We expect General Magic to take on more of the customer interaction layer over time, not by owning more screens, but by owning the moments where customers are stuck, confused, or at risk of dropping out.

Insurance is the starting point because the need is clear and the impact is immediate. The same approach applies anywhere follow through breaks down and coordination becomes costly.

The opportunity is not just operational efficiency. It is a better experience for customers and calmer, more focused teams for insurers.

Parting words

We believe messaging will become a core operating channel for insurance, not as a support tool, but as a place where real work gets done.

When implemented correctly, AI driven messaging reduces friction, recovers lost workflows, and lowers operational load, while making insurance feel more responsive and human at the moments that matter.

A & J

User interfaces have evolved over time, from command lines used by technical operators to graphical interfaces that made insurance software usable at scale. That shift unlocked efficiency, but it also locked companies into portals, menus, and rigid workflows.

Customer behavior has changed since then. When something breaks, when a quote stalls, when a claim feels unclear, people do not want another portal or a long call queue. They want a simple interaction that moves things forward.

Insurance is now feeling the gap between how work is structured internally and how customers expect to engage.

Thesis

We believe the next shift in insurance operations will happen in messaging.

More of the work that currently happens across calls, emails, and portals will move into simple text conversations that can actually complete tasks. Not notifications. Not reminders. Real work.

SMS is the most reliable place to do this. It is where urgency shows up, where customers respond, and where attention already exists when something matters.

General Magic is building infrastructure to make messaging a dependable operating channel for insurers.

The current state

Across the industry, customers are already trying to interact this way. They reply to reminder texts with questions. They send documents over text. They ask for changes outside of office hours.

Most insurance systems are not built to handle that. SMS is treated as a one way notification pipe, while the actual work still requires a call, a portal login, or manual follow up from a team member.

The result is predictable. Quotes go cold. Claims stall. Documents are chased. Call centers absorb avoidable volume. Teams spend time coordinating instead of resolving.

Why this has not worked before

Messaging sounds easy until it has to touch real insurance workflows.

The moment you need identity checks, document collection, policy details, billing changes, endorsements, or exceptions, everything breaks unless the conversation is deeply connected to systems of record.

On top of that, this is not something a team can ship once and forget. Reliability matters. Accuracy matters. One wrong answer at the wrong moment destroys trust.

Most insurers do not want to build and maintain this layer themselves. It requires AI expertise, ongoing iteration, operational discipline, and deep integration work.

The AI messaging layer for insurance

We believe there is a new category here.

Insurers will operate customer facing AI agents that live in messaging and are directly connected to their internal systems. On the company side, the product looks like an operational workspace where each conversation represents a workflow in motion. On the customer side, it becomes a real alternative to calling, waiting, and bouncing between departments.

Some workflows are handled dynamically in simple cases. More complex flows are designed ahead of time and triggered by customer messages or internal events like quotes, FNOL, or renewals.

Why insurance first

Insurance is one of the most procedural industries in existence.

Work is stateful, multi step, and full of edge cases. Human teams act as glue between systems, not because judgment is always required, but because coordination is.

You see it everywhere. Quote to bind. FNOL to adjust to approve to pay. Endorsements. Renewals. Document chasing. Follow ups.

At the same time, insurance already has strong systems of record. Policy systems. Claims systems. Billing. CRMs and agency management systems. The logic exists. The breakdown happens at the point of customer interaction.

Where General Magic fits

We offer a product called Cell, and we apply it at the moments where insurance workflows most often stall.

Cell powers SMS agents that insurers deploy after a quote, at first notice of loss, during servicing, and through renewal. The agent answers questions, collects documents, runs verification, handles KYC flows, and moves workflows forward without another call.

On the insurer side, this is not about replacing teams. It is about reducing manual follow up, smoothing handoffs, and making outcomes more consistent.

What AI adds on top of basic SMS

Basic SMS sends reminders. AI turns reminders into completed actions.

Instead of telling someone to remember an appointment, the agent can confirm or reschedule in chat, handle conflicts immediately, and answer the small questions that cause no shows. That removes ambiguity and friction, which is where most drop off happens.

AI also enables personalization that actually matters. Not surface level fields, but adjustments based on behavior, constraints, language, timing, and intent. That builds trust and improves follow through.

Forms become conversations. Structured data is collected naturally and mapped directly into systems, which means fewer portals, fewer logins, and fewer abandoned flows.

AI provides just in time guidance for procedural work. It breaks tasks into small actions, detects confusion, and explains next steps clearly. That is what reduces delays and “I will do it later” behavior.

When things get complex, escalation works. The agent detects risk, routes to a human with full context, and resumes the workflow after handoff. That is how messaging becomes dependable in real operations.

What needs to be true to win

For messaging to replace calls, it has to be faster than doing the work manually. It has to be reliable enough that teams trust it in high stakes moments. It has to feel native to the insurer’s brand and flexible to individual customer needs.

Most importantly, it has to be connected to systems of record. Messaging without action is just a nicer FAQ.

Time to value also matters. Insurers should see impact in the first conversations, not after months of setup.

Parting words

We believe messaging will become a core operating channel for insurance, not as a support tool, but as a place where real work gets done.

When implemented correctly, AI driven messaging reduces friction, recovers lost workflows, and lowers operational load, while making insurance feel more responsive and human at the moments that matter.

A & J