Security in quantum cryptography [1, 2] is continuously challenged by inventive attacks [3][4][5][6][7] targeting the real components of a cryptographic setup, and duly restored by new countermeasures [8][9][10] to foil them. Due to their high sensitivity and complex design, detectors are the most frequently attacked components. Recently it was shown that two-photon interference [11] from independent light sources can be exploited to avoid the use of detectors at the two ends of the communication channel [12,13]. This new form of detection-safe quantum cryptography, called Measurement-Device-Independent Quantum Key Distribution (MDI-QKD), has been experimentally demonstrated [13][14][15][16][17][18], but with modest delivered key rates.Here we introduce a novel pulsed laser seeding technique to obtain high-visibility interference from gain-switched lasers and thereby perform quantum cryptography without detector vulnerabilities with unprecedented bit rates, in excess of 1 Mb/s. This represents a 2 to 6 orders of magnitude improvement over existing implementations and for the first time promotes the new scheme as a practical resource for quantum secure communications. * marco.lucamarini@crl.toshiba.co.uk arXiv:1509.08137v2 [quant-ph] In Quantum Cryptography, a sender Alice transmits encoded quantum signals to a receiver Bob, who measures them and distils a secret string of bits with the sender via public discussion [1].Ideally, the use of quantum signals guarantees the information-theoretical security of the communication [2]. In practice, however, Quantum Cryptography is implemented with real components, which can deviate from the ideal description. This can be exploited to circumvent the quantum protection if the users are unaware of the problem [19].Usually the most complex components are also the most vulnerable. Therefore the vast majority of the attacks performed so far have targeted Bob's single photon detectors [3][4][5][6][7]. 13] is a recent form of Quantum Cryptography conceived to remove the problem of detector vulnerability. As depicted in Fig. 1(a), two light pulses are independently encoded and sent by Alice and Bob to a central node, Charlie. This is similar to a quantum access network configuration [20], but in MDI-QKD the central node does not need to be trusted and could even attempt to steal information from Alice and Bob. To follow the MDI-QKD protocol, Charlie must let the two light pulses interfere at the beam splitter inside his station and then measure them. The result can disclose the correlation between the bits encoded by the users, but not their actual values, which therefore remain secret. If Charlie violates the protocol and measures the pulses separately, he can learn the absolute values of the bits, but not their correlation. Therefore he cannot announce the correct correlation to the users, who will then unveil his attempt through public discussion.Irrespective of Charlie's choice, the users' apparatuses no longer need a detector and the detection vulnerability of Quantum Cryp...