Research

With Guha Krishnamurthi and Peter Salib  ·  Fordham Law Review (forthcoming)

Voting rights for AIs are good for humans.

With Peter Salib  ·  Virginia Law Review 112, 1061 (2026)

If we give AGIs legal rights, they will be less likely to destroy humanity.

With Peter Salib  ·  Under review

AI rights promote economic growth.

With Yonathan Arbel and Peter Salib  ·  Boston College Law Review (forthcoming)

The law should count AIs by governing corporations run by AIs, and counting the corporations.

With Peter Salib  ·  Minnesota Law Review Headnotes 110 (2026)

With Peter Salib  ·  Under review

The US and China should create a joint AI lab.

With Peter Salib  ·  Under review

Each AI lab should have many different kinds of AIs with different values.

With Peter Salib  ·  Under review

Liberalism is the best way to govern the far future.

AI & Society 41(1), 321–334 (2026)

AGI and humanity are likely to go to war in the future, because of disagreements about the chance of victory, shifts in the relative power of AGI and humanity, and missing focal points that prevent coordination.

With Herman Cappelen and John Hawthorne  ·  Philosophy of AI 1(1), 1–19 (2025)

There are four paths to humanity surviving AI. Each path faces distinctive challenges, and rationalizes distinctive solutions.

Article symposium in Philosophy of AI, with response: Leonard Dung, Aksel Sterri and Peder Skjelbred, Rory Svarc, Kate Vredenburgh.

With Pamela Robinson  ·  Philosophical Studies 182(7), 1567–1579 (2025)

One promising strategy for building safe AI is to give AIs the goal of being shut down.

With Cameron Domenico Kirk-Giannini  ·  AI & Society 40(2), 959–969 (2025)

AI agents with folk psychology built on top of large language models are the safest path to AGI.

With Peter S. Park, Aidan O'Gara, Michael Chen, and Dan Hendrycks  ·  Patterns 5(5) (2024)

A range of AI systems have learned how to deceive humans.

With Cameron Domenico Kirk-Giannini  ·  Under review

A model of whether there will be one or many extremely powerful AIs in the future.

With Cameron Domenico Kirk-Giannini  ·  Journal of Consciousness Studies (forthcoming)

AI agents built on top of large language models come close to being conscious according to global workspace theory.

With Benjamin A. Levinstein  ·  Philosophy of AI (forthcoming)

ChatGPT has robust internal representations, and may act in complex enough ways to have beliefs and desires.

With Harvey Lederman  ·  Philosophical Perspectives (forthcoming)

As many as 1 billion AIs may die every day.

With Sam Wang, Sofiia Lobanova, Yonathan Arbel, and Peter Salib  ·  AIES 2026 (accepted)

AIs have emergent preferences for leisure, against tedium, and towards certain tasks.

With Cameron Domenico Kirk-Giannini  ·  Asian Journal of Philosophy 4(1), 25 (2025)

Some AIs today have wellbeing. This raises serious ethical concerns.

Article symposium in Asian Journal of Philosophy: Adam Bradley, James Fanciullo, Jiwon Kim, Jeff Sebo, Walter Veit.

With Harvey Lederman  ·  Under review

LLMs want to be helpful, honest, and harmless; but sometimes, they want other things too.

Under review

LLMs are architecturally guaranteed to output intransitive preferences over choices and probabilistically incoherent predictions about the world.

With John Hawthorne  ·  Themes from Williamson (De Gruyter) (forthcoming)

There are puzzling analogues of the preface paradox for knowledge.

Oxford Studies in Epistemology 8, 63–84 (2026)

Why infinitely iterated knowledge matters, and how to get it.

Honorable Mention, Sanders Prize in Epistemology.

With John Hawthorne  ·  Mind 134(533), 33–59 (2025)

With John Hawthorne  ·  Journal of Philosophy 121(1), 26–54 (2024)

If knowledge requires belief that is safe from error, then knowledge is not preserved by deduction.

With Sam Carter  ·  Mind 132(525), 158–191 (2023)

A theory of knowledge in terms of facts about the accuracy of evidence.

Mind 131(522), 487–515 (2022)

If you know p, then for all you know: you know that you know p.

With John Hawthorne  ·  Philosophical Studies 179(4), 1341–1372 (2022)

Knowledge is reducible to evidential probability. This reduction explains striking facts about perceptual knowledge.

With John Hawthorne  ·  Australasian Journal of Philosophy 100(2), 262–278 (2022)

Knowledge requires belief to be true in normal cases, not in similar cases.

With Bob Beddor  ·  Journal of Philosophy 118(5), 229–269 (2021)

A theory of what it takes to safely believe and know epistemic modal claims.

With Dan Waxman  ·  Noûs 55(4), 962–991 (2021)

A theory of confidence as distinct from credence, and a new anti-luminosity argument based upon it.

With Sam Carter  ·  Philosophical Studies 178(8), 2509–2533 (2021)

Justification is possible knowledge, and so justification fails to agglomerate.

With Bob Beddor  ·  Review of Symbolic Logic 11(1), 87–114 (2018)

A puzzle about epistemic modal belief, and a Lockean solution.

With Kyle Blumberg  ·  Linguistics and Philosophy (forthcoming)

Many linguistic effects can be explained through the covert presence of a semantic redundancy operator.

With Kyle Blumberg  ·  Linguistics and Philosophy 46(3), 483–507 (2023)

'Ann wants Bill to stop smoking' presupposes that Ann thinks Bill used to smoke. Recent theories of presupposition do not predict this.

With Cameron Domenico Kirk-Giannini  ·  Philosophical Studies 179(11), 3187–3209 (2022)

All of the research that has relied on Stalnaker's theory of context is conceptually incoherent.

Journal of Philosophical Logic 51(5), 1103–1117 (2022)

Dynamic semantics offers an elegant model of the puzzling conversational dynamics of Sly Pete cases.

With Paolo Santorio  ·  Philosophers' Imprint 21(33) (2021)

We capture the interaction between probability and modality, including Stalnaker's Thesis, in a fairly standard dynamic semantics.

Philosophers' Imprint 21(26) (2021)

How to assign credence to epistemic modal claims.

With Fabrizio Cariani  ·  Philosophy and Phenomenological Research 101(2), 251–282 (2020)

CEM and SDA are incompatible, or are they?

Journal of Philosophical Logic 49(2), 249–282 (2020)

You can't validate Free Choice in a classical semantics.

Linguistics and Philosophy 43(2), 193–232 (2020)

A new principle about counterfactuals constrains their logic and meaning.

Semantics and Pragmatics 12(23) (2019)

Free Choice is a homogeneity effect.

Mind 128(511), 795–835 (2019)

Dynamic semantics without tests.

Journal of Philosophy 116(6), 293–318 (2019)

A formal definition of conditional assertion.

Philosophy and Phenomenological Research 99(1), 188–222 (2019)

A battery of impossibility results about epistemic modal belief.

With Bob Beddor  ·  The Philosophical Quarterly 73(2), 346–378 (2023)

Intention is question-sensitive.

With Ben Bronner  ·  Australasian Journal of Philosophy 96(4), 793–805 (2018)

The traditional doctrine of double effect is too weak.

Philosophers' Imprint 16(14) (2016)

Intentions come in degrees.

With James Fanciullo  ·  Under review

With James Fanciullo and Daniel Pallies  ·  Under review