🔍 Read the full analysis: Block’s Dessa Laughs, Hesitates And Interrupts Like A Person. Why A Human-Sounding Voice Is No Longer Proof Of Anything on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Block announced Dessa, a speech-to-speech research model whose selected demos include laughter, interruptions and emotional responses. The model has not been released for public testing, and independent verification, technical details and safety safeguards remain limited.
Block announced Dessa, a speech-to-speech research model, on October 23, showing recordings in which it laughs mid-sentence, responds to emotion and interrupts or backchannels during conversation. The demonstrations raise questions about how natural-sounding AI could affect phone-based fraud, but the model has not been independently tested and is not publicly available.
Block says Dessa was trained from scratch on recorded human conversation, without text or pretrained models. The company describes it as a full-duplex system: it can listen while speaking, detect when to respond and produce features such as laughter, sarcasm and pauses. Block says it runs faster than real time on a MacBook Pro M4 Max or better, but has not provided a latency figure.
The recordings on Block’s project page are the main public evidence so far. They include a conversation about a forgotten film title and a response to an anniversary surprise. Block says the clips are live recordings and were not steered. That claim has not been independently verified, and journalists have not had access to run the model themselves.
Outside coverage has been more qualified than the launch claims. The Verge described some exchanges as “a little awkward and uncanny,” while saying Dessa came across as more natural than another model in a comparison. TechCrunch reported that it was unclear whether emotional delivery was guided by tags. CoinDesk said Block had not disclosed the model’s size, benchmark results or a public release date.
Infographic 01 · Dessa fact sheet
What Block claims, what was heard, what is missing.
Every capability below comes from Block or its selected recordings. Nobody outside Block has run the model yet.
| Topic | What we know | Status |
|---|---|---|
| Training data | Human conversation only: “no text, no pretrained models” | Block’s claim |
| Conversation | Full-duplex: listens while it talks, backchannels, cuts in | Block’s claim |
| Emotion | Hears and speaks laughter, sarcasm, sighs; unclear if steered by tags (TechCrunch) | Block’s claim |
| Speed | “Faster than real time”; no latency figure | Block’s claim |
| Hardware | Trained on one NVIDIA GB200 NVL72; runs on a MacBook Pro M4 Max or better | Block’s claim |
| Demos | Some chats “a little awkward and uncanny”, more natural than GPT-Live-1 (The Verge) | reporters heard it |
| Size, benchmarks, release date | Not disclosed (CoinDesk) | not disclosed |
| Paper, model card, license, weights | None found as of 26 Oct 2026 | not disclosed |
| Languages | Not listed; Block calls learning new languages an open question | not disclosed |
| Voice cloning safeguards | No statement on watermarking, consent checks or imitating a specific person | not disclosed |
Who can use it
research preview
- Public: a waitlist; “launch links are coming soon”
- Early access: “a small group of AI safety researchers”, institutional email required
- Business use: none announced; Square already uses voice AI for ordering (TechCrunch)
Who built it
from Block’s page
- A “(very) small AI research lab at Block” reporting to Jack Dorsey
- Founded by former members of Dessa, the Toronto startup Square acquired in 2020 (CoinDesk)
- Lead Ragavan Thurairatnam: “a GPT3 moment for speech”
Sources: Block, Dessa project page (23 Oct 2026); Ragavan Thurairatnam on X (23 Oct 2026); The Verge (23 Oct 2026); TechCrunch (24 Oct 2026); CoinDesk (24 Oct 2026).
Powered by Thorsten Meyer AI
Natural Voices Raise Fraud Risks
The demonstrations matter because natural timing and emotional cues can make it harder for someone on a call to distinguish a synthetic voice from a person. That does not establish that Dessa can impersonate a specific individual or that it has been used for fraud. It does show why listeners should not treat a convincing voice, by itself, as proof of identity.
The FBI’s 2025 Internet Crime Report, published in April 2026, counted 22,364 complaints mentioning AI and $893 million in reported losses. The report also described more than $5 million in losses from distress scams using cloned voices of loved ones. Those figures cover reported incidents and do not show that Dessa contributed to any of them; the model is not publicly accessible.
For organizations, the practical concern is how to verify requests made over voice calls, especially requests to move money or disclose sensitive information. For families, a familiar voice should not override a request to act urgently. These precautions address voice impersonation generally, not a confirmed use of Block’s model.
Products Worth Considering
LG UltraGear 27-Inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDMI 2.1
Sony WH-1000XM4C Wireless Noise-Cancelling Headphones, Black
Apple AirPods Pro 3 Wireless Earbuds with Active Noise Cancellation
As an affiliate, we earn on qualifying purchases.
What Block Has Disclosed
Block, the company behind Square and Cash App, published the Dessa project page on October 23, 2026. Co-founder Jack Dorsey described talking with the model as like talking with a person, citing emotion, nuance, pauses and natural cadence. Dessa lead Ragavan Thurairatnam called the work “a GPT3 moment for speech.” These are company and team characterizations, not results from an independent evaluation.
Block says the research came from a small AI lab formed by former members of Dessa, a Toronto deep-learning startup acquired by Square in 2020. The company says training used one NVIDIA GB200 NVL72 and “hundreds of times fewer words” than large language models require. It has not published the underlying data, methodology or measurements needed to assess those comparisons.
Access is currently limited to a waitlist and early access for a small group of AI safety researchers, according to the source material. As of October 26, there was no public model, API, technical paper or model card. Block has also left open questions such as how efficiently the system can learn new languages.
“Talking with it feels like talking with a person…full of emotion, nuance, pauses, and a natural cadence.”
— Jack Dorsey, Block co-founder
Testing and Safeguards Remain Open
Independent performance remains unverified. Block has not released the model or published a paper, model card, benchmark results, license or public release date. The company-selected recordings do not establish how Dessa performs across ordinary conversations, different speakers or languages, or under independent testing.
It is also unclear whether the model can imitate a particular person, what controls exist to prevent misuse, and whether its audio is watermarked or otherwise identifiable as synthetic. Block says the model chooses its own voice and personality, but that does not answer how identity imitation is handled. The available information does not establish whether the system has consent checks or other safeguards.
Block’s claims that the clips were live and not steered have not been verified by outsiders. Reporters have also raised the possibility that emotional delivery could be influenced by tags. Until independent access and technical documentation are available, claims about the model’s capabilities should be treated as company-reported descriptions, not settled findings.
Public Access and Evaluation Ahead
The next substantive milestone would be access that lets independent researchers evaluate the model, alongside documentation on its training, performance and safeguards. Block has said a small group of AI safety researchers will receive early access, but has not announced a date for a broader release or a public API.
Until those details emerge, the demos show what Block says the model can do in selected recordings, not how reliably it behaves or how it could be misused. Organizations and families can reduce exposure to voice impersonation now by verifying consequential requests through a separate, trusted channel rather than relying on how familiar a caller sounds.
Key Questions
What is Block’s Dessa?
Dessa is a research speech-to-speech model that Block says can listen and speak at the same time and respond with conversational cues such as pauses, laughter and interruptions.
Can the public try the model?
Not at present, based on information available as of October 26, 2026. Block has described a waitlist and limited early access for AI safety researchers, but has announced no public release date.
Has an independent group verified the demos?
No independent testing has been reported in the source material. The available recordings were selected by Block, and outside reporters have not been able to run the model themselves.
Does Dessa clone a specific person’s voice?
That remains unclear. Block says Dessa chooses its own voice and personality, but has not publicly explained whether it can imitate a particular person or what safeguards would prevent that.
How can people respond to suspicious voice calls?
Do not rely on voice familiarity alone. For urgent requests involving money or sensitive information, hang up and call back using a number already known to be genuine; families can also agree on a private code word.
Source: ThorstenMeyerAI.com
