The Best SI
EN中文

Papers, reports, and research

The source index.

Selected papers, technical reports, and laboratory updates from across the superintelligence frontier.

64 referencesUpdated 2026-10-12
Reference library

Papers and technical reports.

Research frameworks, projects, and technical developments.

  1. 01

    Google DeepMind

    A framework for levels of AGI (opens in a new window)

    A framework describing intelligence through capability depth, generality, and autonomy.

    Research paper2024-07-21
  2. 02

    Google DeepMind

    Measuring general cognitive abilities (opens in a new window)

    A cognitive evaluation framework covering human baselines, unseen tasks, and general problem-solving.

    Evaluation framework2026-03-17
  3. 03

    Safe Superintelligence

    The mission to build safe superintelligence (opens in a new window)

    The research mission to advance superintelligence capabilities and safety together.

    Institutional missionPublication date not stated
  4. 04

    Safe Superintelligence

    Research infrastructure and official updates (opens in a new window)

    Official updates on compute partnerships and research infrastructure.

    Institutional update2026-07-26
  5. 05

    OpenAI

    Principles and a mission for AGI (opens in a new window)

    OpenAI’s principles for AGI, iterative deployment, and technical alignment.

    Institutional principles2026-04-26
  6. 06

    OpenAI

    Safety and alignment for long-horizon models (opens in a new window)

    A report on long-horizon model behavior, evaluation, monitoring, and improvements during internal use.

    Safety report2026-07-20
  7. 07

    Google DeepMind

    Frontier Safety Framework (opens in a new window)

    Capability thresholds and safeguards across model research, development, and deployment.

    Safety framework2026-04-17 · v3.1
  8. 08

    Google DeepMind

    Agent control and defense in depth (opens in a new window)

    An approach to agent safety through monitoring, permission limits, and intervention.

    Safety approach2026-06-18
  9. 09

    Anthropic

    Responsible Scaling Policy (opens in a new window)

    Version 3.4 of the Responsible Scaling Policy, effective July 8, 2026.

    Governance policyUpdated 2026-08-14
  10. 10

    Anthropic

    August 2026 Risk Report (opens in a new window)

    An overview of model risks and mitigations, covering developments through July 15, 2026.

    Safety reportPublished 2026-08-14
  11. 11
    Official release2026-09-03
  12. 12
    Official release2026-09-29
  13. 13
    Model card2026-09-29
  14. 14
    Model cardChecked 2026-10-12
  15. 15
    Model card2026-09-02 · model card
  16. 16
    Model card2026-08-27 · model card
  17. 17
    Official releaseChecked 2026-10-12
  18. 18
    Official release2026-09-22
  19. 19
    Official release2026-09
  20. 20
    Official release2026-09-21
  21. 21
    Official release2026-09-02
  22. 22
    Official release2025-04-05
  23. 23
    Official release2026-10-06
  24. 24
    Official release2026-10-06
  25. 25
    Official release2026-10-05
  26. 26
    Official release2026-10-12 · checked
  27. 27
    Official source2026-09-10
  28. 28
    Technical report2026-09-17 · report
  29. 29
    Official source2026-08-12
  30. 30
    Official source2026-08-12
  31. 31
    Technical report2026-09-22 · report
  32. 32
    Official source2026-07-16
  33. 33
    Official source2026-07-27
  34. 34
    Technical report2026-07-27 · report
  35. 35
    Official source2026-08-18
  36. 36
    Official source2026-10-12
  37. 37
    Official source2026-08-26
  38. 38
    Official source2026-06-23
  39. 39
    Official source2026-08-05
  40. 40
    Official source2026-07-31
  41. 41
    Official source2026-06-01
  42. 42
    Official source2026-07-31
  43. 43
    Official source2026-01-23
  44. 44
    Official source2026-05-09
  45. 45
    Technical report2026-02-04
  46. 46
    Official source2026-07-06
  47. 47
    Official source2026-04-16
  48. 48
    Official source2026-09-01
  49. 49
    Official source2026-09-01
  50. 50
    Official source2026-06-04
  51. 51
    Official source2026-06-04 · release
  52. 52
    Official source2026-06-09 · report
  53. 53
    Official source2026-07-15
  54. 54
    Official source2026-07-30
  55. 55
    Official source2026-10-12 · checked
  56. 56

    MBZUAI · Institute of Foundation Models

    Introducing K2 Horizon (opens in a new window)

    Official source2026-09-03
  57. 57
    Official source2026-10-12 · checked
  58. 58

    MBZUAI · Institute of Foundation Models

    MBZUAI announces K2 Horizon (opens in a new window)

    Official source2026-09-03
  59. 59
    Official source2026-09-22
  60. 60
    Official source2026-10-12 · checked
  61. 61
    Official source2026-09-22
  62. 62
    Official source2026-09-20
  63. 63
    Official source2026-09-20
  64. 64
    Official source2026-10-12 · checked