Runway Characters: Real-time AI avatars

Runway Characters: Real-time AI avatars

A real-time AI avatar is a video-based conversational agent that can listen, respond, and show synchronized facial movement during a live interaction.

Runway Characters is not just another image-to-video feature. It points to a bigger shift: interfaces that talk back, maintain expression, and sit inside websites, apps, support journeys and training environments as an interactive layer.

From chatbot box to embodied interface

For years, the consumer web has treated conversation as a text box. Runway Characters pushes the interaction into a more human-shaped format: a visual character with a voice, a defined personality, domain knowledge and live responsiveness.

The enterprise value is not the avatar; it is the controlled interaction layer around the avatar.

A controlled interaction layer is the set of rules, knowledge sources, permissions, actions, escalation paths and measurement signals that determine what the avatar can say and do.

This is why the product is more interesting for operators than for novelty-watchers. A branded face is easy to demo; turning it into a trusted, scalable and measurable service interface is the hard part.

The mechanism: image, voice, knowledge and action

The mechanism is straightforward: a single reference image defines the character, voice and personality shape the interaction, a knowledge base keeps the response inside a domain, and API actions allow the character to do work rather than just talk.

For enterprise teams, this turns the avatar from a creative asset into a governed service surface that sits between consumers, content, data and workflow.

A governed service surface is a customer-facing interface whose content, permissions, actions, analytics and escalation rules are deliberately controlled.

Because the avatar can combine expression, domain knowledge and actions in the same interaction, the experience can move from navigation to guided execution.

That is the commercial hinge. The avatar is not valuable because it smiles; it is valuable when it helps someone finish a task faster, with less confusion and fewer handoffs.

Where Runway Characters could create real utility

The obvious use cases are the ones Runway highlights: tutoring and education, customer support, training simulations, and interactive entertainment or gaming. Those are credible because the value depends on response, patience, expression and repetition.

The stronger enterprise use case is guided commerce and product selection. A character that understands a product range, asks clarifying questions, checks fit, explains trade-offs and hands off to the right next step could reduce decision friction in categories where consumers need guidance.

Brand and marketing experiences are another useful path, but only if they avoid becoming mascot theatre. A brand character should answer, guide, qualify, educate or convert; otherwise it is just a high-cost animation layer with weak business intent.

The real question is not whether the avatar looks impressive; it is whether the interaction reduces effort, shortens a service path, or improves a decision.

The operating model matters more than the character

The failure mode is predictable: teams launch a polished avatar before defining ownership, content governance, privacy boundaries, escalation logic and measurement. That creates a visible interface with unclear accountability.

For consumer experience platforms, the hard work sits behind the face. The avatar needs approved knowledge, consent-aware data access, clear action limits, analytics events, brand controls, QA scripts and a fallback path when confidence is low.

This also changes the content model. Product information, policy content, service scripts and training material need to be structured enough for a live character to use safely, not just published as static pages for humans to browse.

Runway Characters takeaway for enterprise teams

Runway Characters should be evaluated less like a creative tool and more like a new front-end pattern for service, learning, commerce and brand interaction. The adoption question is not “can we make a character?” but “which consumer or employee journey deserves a live conversational interface, and can we govern it?”

Takeaway: Treat real-time AI avatars as governed service surfaces, not animated brand assets. The winning teams will connect character design to knowledge governance, journey ownership, action permissions, measurement and fallback logic before scaling the experience.

A few fast answers before you act

What is Runway AI?

Runway is an AI company building generative media tools and world-simulation research systems. Runway describes its mission as building AI to simulate the world through the merging of art and science.

What is Runway Characters?

Runway Characters is Runway’s real-time avatar product for creating conversational video characters with customizable appearance, voice, personality, knowledge and actions.

Why does it matter for brands?

It matters because it can turn static content, support flows and training material into live guided interactions that feel more natural than a chatbot.

What are the best first use cases?

The best first use cases are narrow, repeatable journeys where guidance reduces effort: product advice, customer support triage, onboarding, training practice and education.

What is the main enterprise risk?

The main enterprise risk is launching a convincing avatar without clear governance over what it knows, what it can say, what it can do and when it must escalate.

How should teams measure success?

Teams should measure task completion, deflection quality, conversion support, time saved, escalation rate, user satisfaction and the cost of maintaining the knowledge base.

Manus AI: Action Engine for Marketing

Manus AI: Action Engine for Marketing

Manus AI is interesting because it changes the unit of AI adoption in marketing. The useful question is no longer whether AI can write better copy, but whether it can safely execute repeatable marketing work across tools, accounts and output formats.

Vaibhav Sisinty, founder of GrowthSchool, frames the hype in the video, but the useful part is the work pattern: browser shopping, download cleanup, Meta ads analysis, Slack triage, influencer research, prototype building and Telegram-based task handoff.

These are not glamorous use cases. They are the small operational gaps that make marketing teams slower than they should be: extracting data, checking dashboards, comparing options, building lists, scanning messages, formatting outputs and turning loose requests into usable artefacts.

The operating shift: from answer to action

Most marketing teams still use AI as an answer layer. They ask for ideas, summaries, drafts, research angles, prompt variants or campaign copy, and then people still move the work manually through browsers, spreadsheets, CMS workflows, ad platforms, project tools and approval chains.

Manus describes itself as an action engine. An action engine is an AI layer that can plan, execute and package work across tools, rather than only generate recommendations.

The mechanism is straightforward: Manus combines planning, browser operation, connectors, file access, code generation and output packaging, so a marketing request can move from prompt to finished artefact without being manually rebuilt in five separate tools.

For marketing teams, that puts the pressure point on operating model design, not on prompt novelty.

This mechanism matters because execution creates real business value only when the system can reach the right tools, use the right data, follow the right rules and hand back something a team can trust.

The marketing question: control before scale

The real question is whether a marketing organization can give any agent safe enough access, clear enough tasks and strong enough controls to make the output usable.

The stance here is clear: treat Manus as a workbench for bounded execution, not as a replacement for marketing judgment.

In a real marketing stack, that distinction matters because the work crosses content systems, asset libraries, product data, CRM, analytics, ad platforms, consent, identity and approval workflows.

The Meta angle matters, but not as gossip

Manus still presents itself as part of Meta, while recent reporting says China has blocked the acquisition or ordered the transaction unwound. That tension deserves a brief mention, but it should not dominate the argument.

The business signal is not the takeover drama. It is that the market is moving from AI tools that advise marketers to AI systems that can sit closer to actual work.

That is why the Meta connection is relevant: ads, creators, messaging and business pages are workflow surfaces, not just media surfaces.

If an execution agent can sit near those surfaces, the commercial value is not another content generator. The value is shorter distance between insight, action, packaging and follow-up.

Governance decides whether this scales

An agent that can open browsers, read accounts, analyze campaigns, create files, draft replies and ship prototypes is useful only when access rights, approval steps and logs are explicit. Before scaling it, marketing teams need to define which accounts can be touched, which actions are read-only, which outputs require human approval, which data is excluded and which records prove what happened.

Without this, the failure mode is obvious. The agent becomes another shadow workflow, fast enough to bypass controls and persuasive enough to hide weak evidence.

That is also where adoption gets decided. People will not use an agent because it is magical; they will use it because it removes low-value work without making them responsible for invisible risk.

What marketing teams should operationalize

The practical move is not to connect everything at once. Start with bounded, reversible work: campaign monitoring, reporting summaries, initial lists of potential creators and influencers, content calendars, competitive scans, meeting follow-ups, prototype briefs and internal workflow cleanup. These jobs have enough friction to matter, enough structure to test, and low enough downside if a human reviewer stays in the loop.

Takeaway: Offerings like Manus AI are useful for marketing when they are treated as execution layers for controlled workflows, with clear access rules, human approval points, source checks, output QA and measurable time saved.

A few fast answers before you act

What is Manus AI?

Manus AI is a general-purpose AI agent designed to execute tasks, not just answer prompts. In marketing, that means it can support research, reporting, campaign analysis, workflow automation and prototype creation when access and review are controlled.

How is Manus different from ChatGPT or Claude?

ChatGPT and Claude are usually used as reasoning and drafting interfaces. Manus is positioned closer to an execution environment because it can use browser operation, connectors and output generation to turn a request into a finished artefact.

Should marketing teams connect Manus to real accounts?

Not without data governance and security review. Start with read-only access where possible, confirm what data leaves your environment, exclude sensitive customer or employee data, require human approval before external actions, and keep logs for every workflow that affects campaigns, customers or brand assets.

Does the Meta acquisition story change the marketing argument?

Only slightly. The ownership story is unstable, but the operating lesson is stable: AI agents are moving closer to ads, creators, messaging, commerce and business workflows.

What is the best first use case for Manus in marketing?

Start with recurring analysis and packaging work. Weekly campaign summaries, potential creator and influencer lists, competitor scans and meeting-to-action-plan workflows are easier to govern than live publishing or customer-facing execution.

Pepsi: The Recycling Rethink

Pepsi: The Recycling Rethink

Sustainability marketing breaks when the system stays the same

Most sustainability marketing fails when the operating reality does not change, and the message asks consumers to do more while leaving the friction, reward, and moment of action unchanged.

That is exactly the problem here. Special Australia says two out of every three plastic and aluminium containers in Australia still do not get recycled, and Pepsi’s promotion in New South Wales (NSW) only worked because it added a materially better incentive to an existing 10c deposit system in a promotion that ran until 22 November 2025.

The Pepsi example is one of the stronger sustainability ideas in recent memory because it changes the behaviour system, not just the brand message. It also won a Gold Spike in Creative Commerce at Spikes Asia 2026.

Pepsi moved the incentive into the machine

Pepsi worked with TOMRA and the NSW Government-run Return and Earn program to add new code to existing reverse vending machines. A reverse vending machine is an automated kiosk that identifies eligible drink containers and issues the deposit refund. The updated flow let a Pepsi barcode trigger an additional voucher and QR journey on top of the standard 10c return, turning a fixed refund mechanic into a live, brand-specific incentive layer inside an existing public recycling system. Alongside the standard 10c refund, the program also added an A$100,000 bonus prize pool, with rewards ranging from A$100 to A$50,000 for eligible Pepsi containers returned through voucher-printing machines in New South Wales.

In operating terms, this is a physical touchpoint workflow redesign, not a media idea bolted onto recycling.

That distinction matters. The innovation was not the poster, the social edit, or the sustainability language. It was the decision to move the brand intervention into the verified transaction itself, where intent, identity, reward, and action already meet.

The real question is not whether consumers care about recycling. It is whether the system makes the desired action feel worth doing right now.

Because the reward is triggered inside the act itself, the behaviour no longer depends on recall or guilt. It depends on immediate reinforcement.

Why this lands beyond one Pepsi promotion

Award-entry materials published on Lions platform The Work say Pepsi container recycling rose 16% in the first week, that 242,000 people participated after eight weeks, and that the initiative delivered a claimed 37% increase in ROI. The same materials say the code was built for broader rollout, while TOMRA says its reverse vending footprint exceeds 87,000 installations in more than 60 markets.

That is the commercially interesting part. The scarce asset here is not ad inventory. It is installed infrastructure that already sits inside a trusted public behaviour loop.

The lesson for enterprise teams is familiar. You usually get more lift by redesigning the moment architecture than by layering one more awareness burst on top of an unchanged flow.

This is why the idea reads like business-tech translation rather than campaign theatre. Pepsi translated a brand objective into machine logic, barcode recognition, partner coordination, and operational rollout across an existing public system.

It is not infinitely portable. Scale would still depend on program operators, machine access, software control, barcode governance, regulatory approval, fraud prevention, and economics that still work after the novelty wears off.

What enterprise teams should take from Pepsi’s recycling redesign

If you want behaviour change, start by auditing the live touchpoint, not the comms plan. Find the moment where the action is verified, identify what data the system already sees, and then ask whether that data can trigger a better reward, message, or next step without rebuilding the whole stack. What Pepsi and its partners changed was not consumer intent. They changed the structure around the decision.

The takeaway is straightforward: when a habit is stuck, stop spending all your energy on persuasion and redesign the transaction layer where the behaviour actually happens.

A few fast answers before you act

What did Pepsi actually change?

Pepsi did not just run recycling creative around the program. It worked with TOMRA and the Return and Earn system to make Pepsi barcodes trigger an additional voucher and QR-based reward flow inside existing reverse vending machines.

Why is this stronger than a normal sustainability ad?

A normal ad leaves the recycling action unchanged. This idea changed the reward logic at the point of verified behaviour, which gives it more operating value than another awareness message.

Could other brands copy the model?

In principle, yes. Special says the functionality is compatible with TOMRA’s broader machine network, and TOMRA says its reverse vending footprint spans more than 60 markets. Whether another brand could actually deploy it would depend on local program requirements, operator permissions, and commercial logic.

What would stop it scaling?

The main blockers are governance and economics, not creativity. A rollout would need machine access, software control, regulatory approval, barcode integrity, fraud safeguards, and a reward model that still makes sense once expanded.

Did it produce measurable results?

Award-entry materials published on Lions platform The Work say Pepsi container recycling rose 16% in the first week, that 242,000 people participated after eight weeks, and that the initiative delivered a claimed 37% increase in ROI.