Presentations and lectures from conferences and events around the world.
Kevin Owocki sits down with Matthew Monahan (Ma Earth) and Holke Brammer (Hypercerts Foundation) to discuss a bold new experiment: a $1 million Quadratic Funding round designed to bring public goods funding beyond the crypto ecosystem and into the mainstream.
Juan Benet speaks with Konrad Kording of the University of Pennsylvania about the surprising computational power of a single neuron, reading the brain's wiring down to molecular detail, and building compilers and simulations as a path toward understanding — and ultimately simulating — the brain.
Juan Benet speaks with Tom Oxley, co-founder and CEO of Synchron, about the Stentrode — a brain-computer interface that reaches the motor cortex through a blood vessel without opening the skull — restoring independence to people with motor impairment and the path toward higher-channel-count neural interfaces.
Juan Benet speaks with Jacques Carolan, a founding Programme Director at ARIA, about getting breakthrough brain treatments to everyone, how ambitious science programs can accelerate neurotechnology, and what it takes to move hard neurotech ideas toward patient impact.
Juan Benet speaks with Ben Rapoport of Precision Neuroscience about treating paralysis, building scalable brain-computer-interface platforms, and how high-quality neural data can transform both clinical care and computational neuroscience.
Juan Benet speaks with Max Hodak of Science Corporation about restoring sight with retinal implants, biohybrid neural interfaces that grow neurons on silicon, the Science BCI ecosystem, and paths toward expanding human intelligence.
AI was trained on us. How do we value data and pay the people who create it? David Dao, co-founder and Chief Scientist of GainForest, unpacks data valuation, Shapley values as a mechanism for attributing data contributions, and how a green data marketplace could fund conservation while improving AI training pipelines.
Evan Miyazono, founder and CEO of Atlas Computing, discusses how US and Chinese approaches to AI regulation are diverging and what that means for global AI safety. While the US moves toward a largely unregulated model, China requires mandatory evaluations for large-scale AI training runs. The conversation explores hard rules versus flexible governance, what rigorous AI evaluations actually look like, and where the greatest systemic risks lie.
Bolton Bailey, a researcher specializing in post-quantum cryptography, explains how Grover's algorithm could enable a quantum-powered miner to dominate the Bitcoin network, rewrite transaction history, and undermine its consensus mechanism — not by stealing private keys, but by breaking the proof-of-work foundation itself. A deep look at the timeline for quantum threats to cryptographic infrastructure and what post-quantum migration paths exist.
Timour Kosters, co-founder of Edge City, discusses why current institutions are too rigid to adapt to accelerating technological change and why humanity needs new frontiers to flourish. He explores designing governance systems from first principles, the role of experimental pop-up cities as living laboratories for new societal models, and how the world's leading scientists, researchers, and entrepreneurs are using Edge Cities to prototype what comes next.
Keolu Fox, the first Native Hawaiian to earn a PhD in genome sciences, shares a visionary approach to technology inspired by Indigenous knowledge systems like the Hawaiian ahupuaʻa land management system. He applies the wisdom of circular, balanced systems perfected over generations to modern challenges including genome data sovereignty, sustainable data centers, Earth-friendly computation, and ensuring marginalized communities retain control over their own biological data.
Anastasia Gamick, co-founder and COO of Convergent Research, explains how Focused Research Organizations (FROs) address critical scientific and technological bottlenecks that fall between academic research and commercial markets. Examples include E11 Bio for neural mapping at scale and [C]Worthy for carbon sequestration tools. FROs are purpose-built, time-limited entities that can focus on groundbreaking infrastructure problems with long time horizons — work that neither a startup nor a university lab can sustain.
Yuri Deigin (CEO of YouthBio Therapeutics) presents the case for partial cellular reprogramming as a path to reversing biological ageing. He explains how Yamanaka factors can reset the epigenetic clock in cells without causing dedifferentiation, the evidence from mouse studies showing lifespan extension and tissue rejuvenation, and the translation challenges toward human therapies. Deigin also covers the landscape of longevity biotech startups, the role of VitaDAO in funding pre-clinical research, and why decentralised science models are uniquely suited to longevity research with its long time horizons and diffuse IP.
Joscha Bach (LiquidAI, formerly MIT Media Lab and Harvard) lays out a computational theory of consciousness — arguing that mind is best understood as a self-organising model that a system builds of itself and its environment. He examines what it would take for a machine to be conscious, the distinction between sentience and sapience, and why current large language models are neither. A rigorous framework connecting neuroscience, cognitive science, and AI that is essential reading for anyone building at the neurotech–AI intersection.
Ed Boyden, inventor of optogenetics and expansion microscopy at MIT and HHMI, presents cutting-edge techniques for mapping and manipulating brain circuits at nanoscale resolution. He covers how expansion microscopy — swelling brain tissue with superabsorbent polymers to physically magnify it — enables imaging of individual synapses and molecular complexes. He also discusses optogenetic tools for controlling specific neuron populations, and how these technologies together open paths toward whole-brain circuit mapping and the understanding of neural computation.
Sumner Norman (Buck Institute for Research on Aging, formerly Caltech) delivers a comprehensive survey of neurotechnology from early EEG and deep brain stimulation through today's high-density electrode arrays and closed-loop BCI systems. He covers the engineering limits that have shaped the field — spatial resolution, biocompatibility, longevity of implants — and where breakthroughs in materials science, wireless transmission, and machine learning are opening new possibilities. A clear-eyed assessment of what is clinically viable now, what is 5–10 years away, and what remains fundamental science.
We present F3, a finality component designed to bring fast finality to Filecoin. F3 leverages a novel consensus protocol, GossiPBFT, that runs on the side and finalizes tipsets in tens of seconds instead of hours, while laying the basis for verifiable cross-chain bridging.
Henrique presents Janus and Granite, a consensus framework and algorithm to bring fast finality to the Filecoin network.
Most algorithms solving Byzantine agreement that are implemented and used in practice today rely on some form of synchrony assumptions. Especially when used in blockchain systems, where scalability is of great importance, purely asynchronous solutions suffer from high message complexity or strong assumptions that weaken the adversary. In particular, an adaptive adversary (roughly translating to DoS attacks in practice) is difficult to overcome without any synchrony assumptions while also keeping message complexity low. We are exploring an idea targeting exactly such an algorithm. Starting from Bracha’s algorithm for randomized Byzantine agreement with quadratic message complexity, we derive a protocol for multi-valued agreement with sub-quadratic message complexity.
We present a practical construction and implementation of timelock encryption, in which a ciphertext is guaranteed to be decryptable only after some specified time has passed. We employ an existing threshold network, the League of Entropy, implementing threshold BLS in the context of Boneh and Franklin's identity-based encryption (IBE) At present this threshold network broadcasts BLS signatures over each round number, equivalent to the current time interval, and as such can be considered a decentralised key holder periodically publishing private keys for the IBE where identities are the round numbers. A noticeable advantage of this scheme is that only the encryptors and decryptors are required to perform any additional cryptographic operations; the threshold network can remain unaware of the TLE and does not have to change to support the scheme. We also release an open-source implementation of our scheme and a live web page that can be used in production now relying on the existing League of Entropy network acting as a distributed public randomness beacon service using threshold BLS signatures.
The way in which the development and commercialisation of the metaverse unfold may result in an open architecture for collaborative innovation and participation, or in a series of silos tightly controlled by a small number of commercial entities. But does the outcome even matter? And, if so, how can we shape it? We will briefly go over the implications for users, developers, and regulators, and the role that emerging standards might play.
Max Tegmark (MIT, Future of Life Institute) and Juan Benet explore how powerful AI and surveillance technologies could enable authoritarian control at civilisational scale — and what technical and institutional interventions can prevent it. Tegmark draws on his AI safety research and work with the Future of Life Institute to argue that open, privacy-preserving, decentralised infrastructure is a prerequisite for preserving human agency in an AI-enabled world. Direct and practical on the connection between cryptography, open protocols, and the defense of democratic freedoms.
Nick Bostrom (Oxford Future of Humanity Institute, author of Superintelligence) and Juan Benet discuss the most plausible paths to artificial general intelligence, what alignment actually requires technically, and the philosophical status of digital minds. Bostrom examines why the alignment problem is so hard, what orthogonality and instrumental convergence imply for AI risk, and how human-level AI might arrive suddenly via recursive self-improvement. A foundational conversation on the civilisational stakes of AI development.
Adam Marblestone (Convergent Research, Schmidt Futures) and Juan Benet map the neurotechnology landscape and the research infrastructure needed to accelerate it. Marblestone, who served as Chief Strategy Officer at Kernel before co-founding Convergent Research, surveys the current state of BCIs, brain mapping, and NeuroAI — and explains why Focused Research Organizations (FROs) are uniquely suited to tackling the coordination bottlenecks that slow the field. A strategic overview of where the field needs to go and what it will take to get there.
Alfonso de la Rocha, Research Engineer at ConsensusLab, describes the technical details of interplanetary consensus.
Guy Goren, Research Scientist at ConsensusLab, presents a simplified subnet model for blockchain scaling.
Juan Benet, CEO of Protocol Labs, outlines the scaling requirements of web3 and the many applications of IPC.
Marko Vukolic, ConsensusLab lead at Protocol Labs, reviews the work done by ConsensusLab since its creation and its future directions.
Matej Pavlovic, Research Scientist at ConsensusLab, introduces the Mir framework and the work being done at ConsensusLab implementing performant BFT consensus.
Sarah Azouvi, Research Scientist at ConsensusLab, discusses the directions being explored for the future of Filecoin mainnet consensus.
Is decentralising OnlyFans the killer application for web3? Sarah Azouvi, Research Scientist at ConsensusLab, explains why it is.
Vivien Quéma, Research Adviser to ConsensusLab, and Yann Vonlanthen, Summer Fellow at ConsensusLab, explore directions for multi-threaded execution in FVM.
The notion of timelock encryption was first introduced on the Cypherpunks mailing list in 1993 by the founder of the crypto-anarchist movement, Tim May, and to date, while there have been numerous attempts to tackle it, none have been deployed at scale, nor made available to be used in any practical way. We have recently changed this status-quo by releasing a free, open-source tool that can finally achieve the lofty goal of encrypting things that cannot be decrypted until a given time has come in meaningful ways with proper security guarantees. In this talk, we will discover how our practical instantiation is relying on an existing threshold network running to produce distributed randomness, drand. We rely on threshold cryptography and decentralization of trust to exploit the existing League of Entropy in order to achieve our goal. We will first recall what all of this means, we will then proceed to see how these building blocks allow us to deploy a practical timelock service.
Do you worry about responsible disclosure because they could instead exploit the time-to-patch to find you and remove you from the equation? Dead man switches exist for a reason... In this talk we present a new form of vulnerability disclosure relying on timelock encryption of content: where you encrypt a message that cannot be decrypted until a given (future) time. This notion of timelock encryption first surfaced on the Cypherpunks mailing list in 1993 by the crypto-anarchist founder, Tim May, and to date while there have been numerous attempts to tackle it, none have been deployed at scale, nor made available to be used in any useful way. This changes today: we’re releasing a free, open-source tool that achieves this goal with proper security guarantees. We rely on threshold cryptography and decentralization of trust to exploit the existing League of Entropy (that is running a distributed, public, verifiable randomness beacon network) in order to do so. We will first cover what all of these means, we will then see how these building blocks allow us to deploy a responsible disclosure system that guarantees that your report will be fully disclosed after the time-to-patch has elapsed. This system works without any further input from you, unlike the usual Twitter SHA256 commitments to a file on your computer.
We will introduce a set of measurement methodologies that allow us to uncover the characteristics and interactions in the public IPFS network. After a big measurement campaign, we reveal presence of IPFS peers in more than 2700 Autonomous Systems and 152 countries, the majority of which operate outside large central cloud providers like Amazon or Azure. We further evaluate IPFS performance, showing that both publication and retrieval delays are acceptable for a wide range of use cases.
drand is an opensource project allowing anybody to run a “randomness beacon”. Its goal? Providing a trustable, verifiable source of public randomness that would enable full transparency in online lotteries, leader election or blockchain smart contracts. This talk is about what distributed randomness is, what it means for developers, and users, and why you’d want to use it. I will also present to you the current ecosystem around drand, and what it enables you to do differently and why it is desirable in a distributed, decentralized web to have public, verifiable randomness. Don’t worry though: we will first go through an easy overview of how it works without diving too much into the gory cryptographic details. In addition, I’ll demo how drand works in practice, and explain you how you can easily use it in your applications since drand nodes can be queried by anybody.
Will Scott talks about the options and opportunities for private access to decentralized data, and announces Protocol Labs private data retrieval fund. Targeted at research and prototyping projects, this funds aims to develop additional protocol-compatible primitives allowing users to interact privately with Web3 content. Check out related posts on our blog.
We present the Filecoin Hierarchical Consensus framework, which aims to overcome the throughput challenges of blockchain consensus by horizontally scaling the network. Unlike traditional sharding designs, based on partitioning the state of the network, our solution centers on the concept of subnets –which are organized hierarchically– and can be spawned on-demand to manage new state. Child subnets are firewalled from parent subnets, have their own specific policies, and run a different consensus algorithm, increasing the network capacity and enabling new applications. Moreover, they benefit from the security of parent subnets by periodically checkpointing state. In this paper, we introduce the overall system architecture, our detailed designs for cross-net transaction handling, and the open questions that we are still exploring.
Alfonso de la Rocha provides the latest updates on ConsensusLab's project B1, focused on hierarchical consensus.
Denis Kolegov provides the latest updates on ConsensusLab's project Y5, focused on implementing Tendermint as a subnet consensus mechanism.
Sarah Azouvi provides the latest updates on ConsensusLab's project B2, focused on preventing long-range attacks by checkpointing state onto the Bitcoin chain.
Vivien Quéma provides the latest updates on ConsensusLab's project G1, focused on the parallelisation of execution in WASM order-execute models.
Alfonso de la Rocha presents his work on Hierarchical Consensus, whose goal is to develop a framework to enable on-demand horizontal scalability of Filecoin (and other blockchain networks) through an alternative sharding/side-chain scheme that supports heterogeneous consensus algorithms in each subnetwork.
Juan Benet, founder of Protocol Labs, talks about the importance of consensus research, some of the challenges ahead, and what PL is doing to address them.
Marko Vukolic, ConsensusLab lead at Protocol Labs, discusses the vision for the lab and some key challenges for decentralization.
Long-range attacks on PoS blockchains and the related “I still work here” attack in BFT-based blockchains are among the major security issues in these systems. Recently, an approach has been proposed to deal with these attacks by anchoring BFT/PoS membership into Ethereum’s proof-of-work blockchain. As Ethereum is abandoning PoW, this approach is no longer viable. This project aims at achieving the same goals using Bitcoin’s PoW.
Single Secret Leader Elections have recently been proposed as an improved leader election mechanism for proof-of-stake blockchains. However, the security gain they provide has not been quantified. In this talk, Protocol Labs research Sarah Azouvi presents a comparison of proof-of-stake longest-chain protocols that are based on Single Secret Leader Elections (SSLE)– that elect exactly one leader per round – versus those based on Probabilistic Leader Elections (PLE) – where one leader is elected on expectation.
Protocol Labs R&D Engineer Alfonso de la Rocha discusses IPFS-FAN, a general architecture of a decentralised network for general-purpose and permissionless computation based on content-addressing.
Research scientist Zixuan (ZX) Zhang gives a workshop on first principles thinking and the history of finance and Filecoin as part of ETHGlobal's HackMoney 2021.
Research Scientist Zixuan (ZX) Zhang does a deep dive to highlight what the Filecoin HyperDrive upgrade enables and how to tactically take advantage of the new change in different chain utilization scenarios.
Protocol Labs R&D Engineer Friedel Ziegelmayer explains the technical accomplishments for solving scaling issues with the rising use of Filecoin.
Protocol Labs R&Der Sarah Azouvi presents her work on game theoretic models of security for cryptocurrencies. This work was originally presented at the Cryptoeconomic Systems Conference at MIT in March 2020.
ResNetLab Research Engineer Alfonso de la Rocha discusses his 'Beyond Bitswap' project. Check out related posts on our blog.
Protocol Labs R&Der Sarah Azouvi presents her research on Winkle, which protects any validator-based byzantine fault tolerant consensus mechanisms, such as those used in modern Proof-of-Stake blockchains, against long-range attacks where old validators’ signature keys get compromised.
Protocol Labs cryptography researcher Anca Nitulescu discusses her work on verifiable computation on encrypted data.
Joel Gustafson talks about what knowledge graphs are, generally, with special attention to his research project, The Underlay, which is a distributed public knowledge graph.
Alan Ransil discusses the Inter-Cranial Abstraction system (ICAS), a ‘Merkle Forest’ of content-addressed types and programs.
Colin Evran and Zixuan Zhang discuss a model of Filecoin circulating supply.
CryptoEconLab researcher ZX leads viewers on a deep dive into the Filecoin economy.
Michael Hammersley talks about enabling transactive energy on legacy power grids. This work was done in conjunction with Alan Ransil, who is associated with both Protocol Labs and the MIT Energy Initiative, as well as Frank O’Sullivan, who is also at the MIT Energy Initative.
Colin Evran and Zixuan Zhang lead Chain Reaction Podcast host Tom Shaughnessy on a technical deep dive into the cryptoeconomics of the Filecoin network.
CryptoEconLab research scientist ZX discusses the mechanisms and models presented in the report "Engineering Filecoin's Economy".
Vector commitments with subvector openings (SVC) allow one to open a committed vector at a set of positions with an opening of size independent of both the vector's length and the number of opened positions. We continue the study of SVC with two goals in mind: improving their efficiency and making them more suitable to decentralized settings. We address both problems by proposing a new notion for VC that we call incremental aggregation and that allows one to merge openings in a succinct way an unbounded number of times. This property leads to faster generation of openings via preprocessing and a method to generate openings in a distributed way. We then proceed to realize SVC with incremental aggregation. We provide two constructions in groups of unknown order. The first one, similarly to that of Boneh et al. (which supports only one-hop aggregation), has constant-size public parameters, commitments and openings. As an additional feature for this construction we propose efficient arguments of knowledge of subvector openings which immediately yields a keyless proof of storage with compact proofs. For our second construction, we propose an incremental aggregation method for the SVC of Lai-Malavolta; this has linear-size parameters but faster openings. Finally, we address a problem closely related to that of SVC: storing a file efficiently in completely decentralized networks. We introduce and construct verifiable decentralized storage (VDS), a cryptographic primitive that allows to check the integrity of a file stored by a network of nodes in a distributed and decentralized way. Our VDS constructions rely on our new vector commitment techniques.
Protocol Labs researcher Michael Hammersley discusses the current state of his work on a decentralized energy grid, a shared project with Alan Ransil, another researcher at Protocol Labs.
A talk by Nicolas Gailly, Research Scientist at Protocol Labs, March 4, 2020
David Dalrymple (Protocol Labs R&D) talks about Dioptics at SYCO 5.
Brian Vohaska (Filecoin Research) talks about SSLE at ConsensusDay 1.
Evan Miyazono (Protocol Labs R&D) talks about mergeable consensus at ConsensusDay 1.
Henri Stern (Filecoin Research) talks about expected consensus at ConsensusDay 1.
Juan Benet (Protocol Labs CEO) talks about consensus hierarchies at ConsensusDay 1.
Juan Benet (Protocol Labs CEO) talks about new directions in consensus at ConsensusDay 1.
In order to decarbonize the power grid, we will make increasing use of resources at the grid edge such as wind, solar and batteries. This implies some significant changes in the structure of the power grid, which will have to transform from its current radial structure into a more distributed one. This talk briefly summarizes the way that the grid is structured now and how electricity pricing is done in transmission networks, and describes how we might extend these mechanisms to support clean energy.
Ben Fisch (Filecoin Research) talks about PoReps at Lab Day 2018.
David Dalrymple (Protocol Labs R&D) talks about IPLD research at Lab Day 2018.
The Underlay is a global, distributed graph database of public knowledge, started at the MIT Knowledge Futures Group in collaboration with Protocol Labs. It aims to provide open access to the content and provenance of all structured data.
An overview of Protocol Labs R&D from team lead Evan Miyazono.
Nicola Greco (Filecoin Research) talks about SNARKs at Zcon0.
Research team lead Evan Miyazono provides an overview of the PL approach to external research, and describes PL Research's Request For Proposals (RFP) program.
Jeromy Johnson (Filecoin Research) talks about VDFs and Filecoin at VDF Day.