Article

Provably Safe Artificial General Intelligence via Interactive Proofs

2021-09-21

Abstract excerpt

<h4>Methods: </h4> are currently lacking to prove artificial general intelligence (AGI) safety. An AGI ‘hard takeoff’ is possible, in which first generation AGI1 rapidly triggers a succession of more powerful AGIn that differ dramatically in their computational capabilities (AGIn≪AGIn+1). No proof exists that AGI will benefit humans or of a sound value-alignment method. Numerous paths toward human exti...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: ff194753-c6be-5e90-a40c-d57e1e5f3a8f
DOI: 10.20944/preprints202109.0358.v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Provably Safe Artificial General Intelligence via Interactive ProofsDOI 10.20944/preprints202109.0358.v1Read

Provable AI Ethics and Explainability in Next-Generation Medical and Educational AI agents: Trustworthy Ethical FirewallSemantic proximity Read

Safe Artificial General Intelligence via Distributed Ledger TechnologySemantic proximity Read

AI Safety and General Collective IntelligenceSemantic proximity Read

A Comprehensive Solution for the Safety and Controllability of Artificial SuperintelligenceSemantic proximity Read

Navigating Artificial General Intelligence (AGI): Societal Implications, Ethical Considerations, and Governance StrategiesSemantic proximity Read

The Symbiosis of Formal Methods and Artificial IntelligenceSemantic proximity Read

The Symbiosis of Formal Methods and Artificial IntelligenceSemantic proximity Read

[Praise of complexity].Semantic proximity Read

The Path to Artificial General Intelligence Through the Concept of a Digital OrganismSemantic proximity Read

Reason Without Reverence: AI as a Partner in Scientific ValidationSemantic proximity Read

Reason Without Reverence: AI as a Partner in Scientific ValidationSemantic proximity Read

A Concern-Oriented Governance Approach for Securing Agentic AISemantic proximity Read