Lovo Bankruptcy + Vapi Outages: Voice Infrastructure Is Risk

On May 27, 2026, Lovo AI filed for Chapter 7 bankruptcy, less than three weeks before scheduled motion-to-dismiss arguments in a class-action lawsuit brought by voice actors Paul Lehrman and Linnea Sage, who alleged their voices were cloned without consent from 2019 and 2020 recordings. The lawsuit had survived partial dismissal. Rather than continue to the next hearing, the company liquidated. Chapter 7 means it is over.

In the same 11-day stretch, Vapi logged three separate outages: a database outage on May 20 that lasted 4 hours and 37 minutes, a degraded service event on May 26 lasting 38 minutes, and increased call failure rates on May 28 lasting 54 minutes. Vapi's headline uptime number is 99.785 percent for the period. That sounds fine until you do the math: 0.2 percent of a month is approximately 86 minutes of downtime. Vapi used more than that in a single incident.

If you are running an AI voice agency and either of these hit you directly, this post is the playbook. If they did not hit you directly, this post is still the playbook. The structural risk does not require you to have been a Lovo customer or to have had a call drop during the May 20 window. It just requires that you are building a client-facing business on infrastructure you do not control.

Why this matters for AI voice agencies

Lovo's bankruptcy is not primarily a story about IP risk or voice actor lawsuits, though those are real. It is a story about what happens when a key upstream vendor in your stack disappears without warning. Agencies that white-labeled Lovo voices now have client deliverables tied to a defunct company. No support. No API continuity guarantee. No contractual obligation to keep anything running. The product they sold their clients is now dependent on infrastructure that is being wound down by a bankruptcy trustee.

That scenario does not require a lawsuit to happen to the next vendor. It requires a funding round that falls through, a pivot that kills a product line, or an acquisition where the acquirer shuts down the legacy API. These are not rare events in the AI tooling market. They are the baseline.

The Vapi situation is a different category of risk, but it compounds the same way. Three outages in 11 days during peak campaign season means missed calls that do not reschedule themselves, missed leads that went cold before a callback, and clients asking why their appointment confirmation campaigns had gaps. Vapi is a well-funded, technically credible infrastructure provider. They still went down three times in 11 days. The issue is not Vapi's competence. The issue is that single-vendor dependency means you inherit 100 percent of that vendor's bad days.

Put both signals together: an AI voice company goes bankrupt mid-lawsuit, and the dominant API provider had its worst reliability month of 2026. The agencies most exposed are the ones who built their entire client delivery stack on a single upstream provider with no fallback. That is a large percentage of the AI voice agency market right now.

What we are doing at Hermes about it

Hermes is an operating platform for AI voice agencies, not an API wrapper on top of a single provider. The distinction matters today more than it did last week.

The infrastructure architecture at Hermes is designed around the assumption that any single upstream provider will have bad days, bad months, or in extreme cases, no days at all. A Vapi degradation should not kill your client's campaign. A provider going dark should not require you to rebuild your entire agent stack. The platform handles provider-layer routing so your agency's uptime is not directly tied to any one vendor's status page.

On the Lovo side specifically: Hermes agencies were never dependent on Lovo as a voice generation layer. But the broader point stands. When you deploy under your brand through Hermes, your clients are insulated from the names of the infrastructure you run on. They see your brand. They interact with your white-labeled workspace. If Hermes ever needed to swap a provider in the stack, that change happens at the platform level. Agencies do not have to rebuild. Clients do not notice. That is the practical value of a platform layer over a raw API integration.

On pricing: Starter is $149 per month with 300 included minutes, Business is $399 per month with 1,000 included minutes, Agency is $699 per month with 2,000 included minutes. Overage is $0.24 per minute. The infrastructure redundancy is not a premium tier add-on. It is how the platform is built.

Action steps for agencies this week

Audit your upstream dependencies before your clients do. Write down every vendor that sits between your agency and your client. Voice AI provider, telephony layer, CRM, automation tool, billing system. For each one, ask: if this vendor went offline for 4 hours or shut down entirely, what breaks immediately for a client? If the answer is "the whole thing," you have a single point of failure. That is the problem to solve before a vendor makes the decision for you.
Contact any clients who noticed gaps during the May 20 Vapi window. If you run campaigns on Vapi and had a client whose outbound had gaps between roughly 9am and 2pm Eastern on May 20, reach out proactively. Do not wait for them to bring it up on the next check-in. A brief message acknowledging the infrastructure event and confirming what you have done to reduce future exposure closes the trust gap before it opens. Silence after an outage reads as not noticing. Proactive acknowledgment reads as operational maturity.
Do not white-label voices from a single TTS vendor without a fallback. Lovo's bankruptcy is an extreme case, but the underlying dynamic is not. Voice AI vendors have raised hundreds of millions of dollars and are still making aggressive product pivots. ElevenLabs is the dominant TTS layer right now and is well-funded at a $11 billion valuation. That does not mean the API terms do not change, that pricing stays stable, or that the product roadmap does not evolve in a direction that breaks your implementation. Build your client deliverables on a platform that abstracts the TTS layer, not on a direct API integration you will have to rebuild when terms change.
Use this week's news as a qualification signal in new sales conversations. Prospects who are serious about building a durable AI voice agency business will care about infrastructure stability when you present the Lovo and Vapi situations concretely. Prospects who are not serious will not ask follow-up questions. The news gives you a natural setup: "You may have seen that Lovo filed for bankruptcy this week and Vapi had three outages in May. Here is how we think about vendor risk and what that means for how we built our delivery stack." The agencies who close the highest-quality clients right now are the ones who can articulate a credible answer to that question.
Move your client delivery infrastructure to a platform layer if you are still running raw API integrations. A direct Vapi or Retell integration gives you low-level control and low prices per minute. It also means you maintain the integration, you handle provider outages, you absorb the full blast radius of any upstream reliability event, and you rebuild when the API terms change. A platform layer like Hermes costs more per month than a raw API key, but it means your agency's capacity to serve clients is not a function of whether any single vendor's on-call engineer is awake at 3am on a Tuesday. At 3 to 10 clients, the math on raw APIs can still work. At 10 to 20 clients, the operational overhead of maintaining raw integrations starts competing with the work of actually growing the agency.

Frequently asked questions

What happened to Lovo AI and does it affect my agency?

Lovo AI, an AI voice generation company, filed for Chapter 7 bankruptcy on May 27, 2026. The filing came less than three weeks before scheduled motion-to-dismiss arguments in a class-action lawsuit brought by voice actors who alleged their voices were cloned without consent from 2019 and 2020 recordings. Chapter 7 means liquidation, not reorganization. The company is done. If your agency white-labeled Lovo voices for clients or used Lovo as part of your voice generation stack, those voices are now backed by a dead company with no ongoing support, no contractual obligations, and no recourse if the API goes dark. The immediate practical impact is disruption for any agency that relied on Lovo-generated voices in active campaigns. The longer-term lesson is about what happens when you build client deliverables on a vendor with no financial floor.

How much revenue does a Vapi outage actually cost an AI voice agency?

It depends on your call volume and how outbound campaigns are structured. Here is a concrete example. An agency running 20 clients at an average of 500 minutes per client per month is processing about 10,000 minutes monthly. That is roughly 333 minutes per day across all clients. Vapi's May 20 outage lasted 4 hours and 37 minutes, which is about 19 percent of a business day. If your campaigns run during business hours only, you lost roughly 19 percent of the day's call capacity. For an agency charging $2,000 per client per month on average, 20 clients represents $40,000 in monthly revenue. A 4.5-hour outage during a business day that represents roughly 6 percent of the month's business hours (assuming 8-hour days, 22 working days) represents about $2,400 in at-risk revenue. That is not catastrophic. But it compounds. Three outages in 11 days, combined with the fact that outbound campaigns are time-sensitive for lead response, means real missed conversions that are unrecoverable.

What is multi-provider orchestration and why does it matter for agencies?

Multi-provider orchestration means your voice platform does not send every call through a single upstream vendor. Instead, the platform maintains relationships with multiple voice AI providers and can route calls through the healthiest available option at any given time. The practical result is that a Vapi degradation, a Retell maintenance window, or a provider going dark does not kill your agency's campaign performance. It is the same principle behind why airlines do not operate on a single engine. The redundancy is not glamorous to describe in a marketing pitch, but it is the difference between losing a client because their outbound went dark for four hours and smoothly finishing their campaign while your infrastructure handled the routing behind the scenes. Single-provider stacks inherit the full risk profile of that one vendor. Platform-level orchestration means the vendor's bad day does not become your client's bad week.

The bottom line

Two infrastructure events in one week are not a coincidence and they are not a warning shot. They are the normal operating environment for AI voice agencies in 2026. Vendors raise money, pivot, get acquired, go dark, and have bad reliability months. The agencies that scale past 10 clients are the ones who treated that as a design constraint, not an edge case.

You cannot control whether Vapi has a database outage on a Tuesday. You can control whether your agency's revenue is the same before and after that outage.

By builders, for builders. One platform. Your brand. Your margins. From $149 per month.

Sources: MLex, Lovo Chapter 7 filing, May 2026 · Law360, Lovo bankruptcy coverage, May 2026 · Vapi status page, May 2026 incident log · IsDown, Vapi uptime tracking May 2026 · TechFundingNews, Synthflow Series A, 2026

Lovo AI Just Filed for Bankruptcy Mid-Lawsuit. Vapi Went Down Three Times in May. Here Is What AI Voice Agencies Need to Do Right Now.