Surfshark has introduced Dausos, a proprietary VPN protocol it says was built specifically for consumer use rather than adapted from enterprise or general networking tools. The release matters because VPN performance is often constrained by protocol design: stronger privacy can add latency, and faster connections can come with trade-offs in resilience or transparency.
For now, Dausos is available in beta on macOS through a separate App Store app. Surfshark says the protocol can deliver speeds up to 30% faster than alternatives including WireGuard, while also tightening traffic separation between users.
Why Surfshark is trying to build its own protocol
Most VPN users never interact directly with the technology carrying their encrypted traffic, but the protocol underneath shapes nearly everything: connection stability, speed, battery use, and how securely sessions are handled. WireGuard, OpenVPN, and IKEv2 dominate the market because they are mature and widely trusted, yet they were not all conceived with mainstream consumer VPN services as their primary use case.
Surfshark’s argument is that consumer VPNs now operate at a scale and pattern of use that justify more specialized engineering. Dausos is meant to reflect that shift. Its core distinction, according to the company, is a dedicated traffic tunnel for each user rather than a shared tunnel handling multiple users at once. In practical terms, Surfshark says this reduces interference between users’ traffic and lowers the chance that one person’s heavy load slows another’s packets.
What is technically different — and what remains unproven
Dausos uses AEGIS-256X2 rather than the more common AES-GCM. That is a notable choice. Encryption design affects not just security but efficiency on modern hardware, and newer constructions can outperform older standards in some conditions. Surfshark also frames AEGIS-256X2 as a post-quantum-minded improvement, though claims around future quantum resistance deserve careful scrutiny across the industry, especially before long-term cryptographic consensus forms.
Any new protocol faces a basic problem: trust has to be earned. That is why external review matters more than marketing language. Surfshark commissioned Cure53 to assess Dausos, and the auditor reported no Critical or High severity findings within the protocol itself, while also saying the company addressed most issues quickly after testing. That is a useful signal, but it is not the same as years of public scrutiny. Open standards such as WireGuard benefit from broad inspection by researchers and implementers over time; proprietary protocols start with a credibility gap and must close it through testing, transparency, and stable real-world performance.
Early results suggest a mixed but credible start
The launch was not frictionless. Some early testers reported weak performance on residential connections, though Surfshark says it deployed a fix quickly. That sequence is familiar in networking products: behavior that looks strong in controlled environments can expose bottlenecks once it meets the variability of home broadband, Wi-Fi congestion, and different ISP routing paths.
Initial hands-on testing described in the launch coverage suggests Dausos now performs competently for everyday browsing and streaming. The reported speed snapshot was nuanced rather than flattering by default: WireGuard retained download performance better, while Dausos held a slight edge on upload speeds. That kind of result is plausible. VPN speed is highly sensitive to geography, server load, peering arrangements, local hardware, and the quality of the base connection, so any single benchmark should be read as indicative, not definitive.
What Dausos could mean for VPN users
If Dausos matures, its strongest appeal may not be raw speed alone but traffic isolation. Consumer VPN services often market privacy in broad terms, yet architecture matters. Giving each user a separate tunnel could help reduce cross-user contention and simplify some security assumptions inside the service. For people who use VPNs daily for public Wi‑Fi protection, routine privacy, or streaming, improvements at the protocol layer can be more meaningful than cosmetic app changes.
The larger question is whether Surfshark will eventually open-source Dausos. The company says that will not happen during the early launch phase, though it is under discussion. That decision may shape how seriously the wider security community takes the protocol. In VPNs, performance claims can attract attention, but long-term trust usually rests on something less glamorous: whether outsiders can inspect the code, test the assumptions, and verify that the protocol behaves as promised.