Transcriptly

What Ancestral Intelligence Can Teach Us About AI | Nanjira Sambuli | TED

TED07:43Added Aug 30, 2026

There's a common African proverb: "When elephants fight, it's the grass that suffers." Policy researcher Nanjira Sambuli says we must apply this thinking to today's AI evolution, asking: When tech giants battle for dominance, who gets trampled in the process? She introduces a new ethical compass for AI, showing how people across the continent are charting a different path for the future of tech. (Recorded at TED2025 on April 9, 2025)

Watch on YouTube →
Contributed by 刘嘉琪

Transcript

Transcript format
  1. 00:04What can the African savannah teach us about AI?
  2. 00:08Take this journey with me.
  3. 00:10Across Africa,
  4. 00:12proverbs are a cornerstone of the oral tradition
  5. 00:15through which Indigenous knowledge and wisdom
  6. 00:17has been passed down from generation to generation.
  7. 00:22The Yoruba people of Nigeria say that a proverb is the horse
  8. 00:26that can carry one swiftly to the discovery of ideas.
  9. 00:31One of my favorite proverbs says, when elephants fight,
  10. 00:34it's the grass that suffers.
  11. 00:36And it has guided me in making sense of much in the world today,
  12. 00:40especially with all the developments in AI.
  13. 00:43The elephants can symbolize great powers,
  14. 00:45be they nation states, corporations,
  15. 00:48broligarchies,
  16. 00:50while the grass comprises people,
  17. 00:52geographies and ecologies
  18. 00:55considered resources to exploit,
  19. 00:57wastelands or charity cases.
  20. 01:00Great power competition, as we're living through today,
  21. 01:03emphasizes the power of the metaphorical elephants
  22. 01:06in their quest for dominance over resources, ideas and innovations.
  23. 01:11In the case of Africa,
  24. 01:13not only are we caught in the middle,
  25. 01:14but we comprise a key battleground
  26. 01:16over which elephants are fighting
  27. 01:18over natural and human resources to power the intelligent age.
  28. 01:23Meanwhile, our perspectives and ambitions
  29. 01:26tend to be drowned out by the din of the elephants.
  30. 01:30But what can we discover when we look past the elephants fighting
  31. 01:34and zoom in on the grass?
  32. 01:36Well, for one, we can learn a simple yet profound concept:
  33. 01:40I am because you are.
  34. 01:44This sums up a set of value systems
  35. 01:46that emerged among the Bantu people of Africa,
  36. 01:49and is known as “utu” in Eastern Africa
  37. 01:52and “ubuntu” in South Africa.
  38. 01:55Utu and ubuntu instill in us a profound appreciation of humanity
  39. 02:00as a quality we owe one another.
  40. 02:03And it's not just about the relationship between humans.
  41. 02:06Ubuntu is also about our relationship with nature and the spiritual or cosmic.
  42. 02:12Ubuntu reminds us we ought to be developing technologies like AI
  43. 02:16for the benefit of all of humanity and our ecology.
  44. 02:21Across Africa, we have exciting examples of embracing the wisdom of ubuntu
  45. 02:26to inform data governance, AI product design and community building.
  46. 02:31I call it ubuntech.
  47. 02:34(Laughter)
  48. 02:36AI development today treats data as if it's an abundant natural resource,
  49. 02:40and we hear this in sayings like, “data is the new oil.”
  50. 02:44But already the limits of this paradigm are being realized,
  51. 02:46as researchers have been sounding the alarm
  52. 02:49that high-quality data to train AI models is drying up.
  53. 02:53Through ubuntu, we conceptualize data differently,
  54. 02:57and we appreciate that it represents lives, cultures and communities,
  55. 03:02so that data governance for us is about the meaningful participation,
  56. 03:07informed consent, self-determination
  57. 03:11and community ownership of data sets from which language,
  58. 03:14nature-based knowledge and Indigenous wisdom are derived.
  59. 03:19(Applause)
  60. 03:20Thank you.
  61. 03:22And this has inspired a concept like data justice
  62. 03:25in our African policy frameworks.
  63. 03:27Data justice matters
  64. 03:29because it means that women, rural women in Africa, for example,
  65. 03:34who possess unique knowledge about agriculture,
  66. 03:39food production, medicine
  67. 03:41and environmental protection
  68. 03:43are represented and visible in data systems
  69. 03:46and agritech solutions.
  70. 03:49Then, when we've been told ours are low-resource languages,
  71. 03:52we're resourcing our languages.
  72. 03:55Conventional AI wisdom demands large language models,
  73. 03:59but African practitioners are making do with little language models.
  74. 04:04See, driven by efficiency as a core value and inspired by our relation to nature,
  75. 04:10initiatives like Lelapa AI have developed lightweight African language models
  76. 04:15that are serving our communities without requiring extensive resources.
  77. 04:20Their Inkuba small language model has been inspired by the dung beetle,
  78. 04:26which can roll up to 250 times its body weight.
  79. 04:29It's small but mighty.
  80. 04:33Inkuba is trained on 0.4 billion parameters
  81. 04:36and outperforms larger models in sentiment analysis
  82. 04:40and displays remarkable consistency across multiple languages.
  83. 04:46We're also building collaborative AI communities.
  84. 04:49We have Masakhane, or building together,
  85. 04:52across over 30 African countries,
  86. 04:55to strengthen natural language processing in our research.
  87. 04:58And this grassroots organizing approach has set out to demonstrate
  88. 05:02that low-resourceness of languages is not a data problem,
  89. 05:05but a societal one best solved through participation.
  90. 05:10In fact, Masakhane have developed a non-traditional authorship model
  91. 05:15that acknowledges and includes all contributors in published papers.
  92. 05:19Be it that you contributed data, lived experience, coded software
  93. 05:24or coordinated research participation.
  94. 05:27And this way, they've been able to publish translation results
  95. 05:30for over 38 African languages.
  96. 05:34These examples of ubuntech
  97. 05:36are our way of charting an alternative path
  98. 05:39to developing and deploying AI solutions in Africa
  99. 05:42by Africa and beneficial for Africa.
  100. 05:47Ubuntech matters for a number of reasons.
  101. 05:49For one, we are asserting agency
  102. 05:52to conceive and build AI futures
  103. 05:54beyond just the ambition of the metaphorical elephants
  104. 05:58and to contribute to a global commons.
  105. 06:01It also allows us to bring forward the Indigenous wisdom of our ancestors
  106. 06:06so that ubuntech is artificial intelligence
  107. 06:09powered by ancestral intelligence.
  108. 06:13(Applause)
  109. 06:18And we also get to remind the elephants,
  110. 06:21they may fight and wield their might all they want,
  111. 06:24but they are also bound to suffer
  112. 06:26if they trample upon the grass
  113. 06:28to the point of irreparable damage or extinction.
  114. 06:31But when their power is exercised in relation to others,
  115. 06:35it makes them ecosystem engineers
  116. 06:37and redirects their energy towards helping sustain a healthy
  117. 06:41and beneficial environment for everyone.
  118. 06:44And we see this in the savanna.
  119. 06:46When African elephants leverage their power to trample upon dense shrubs
  120. 06:50and acacia trees,
  121. 06:52they make room for smaller species to coexist.
  122. 06:55When they disperse seeds as they trek across the land,
  123. 06:58they help generate new growth
  124. 07:00and maintain the biodiversity of the savanna ecosystem.
  125. 07:03So in relationality, in coexistence,
  126. 07:07the power of the elephants is majestic.
  127. 07:10And it's a life force for themselves, other wildlife,
  128. 07:13and the savanna ecosystem.
  129. 07:16I believe if we can reimagine humanity
  130. 07:19beyond just the ambition of the metaphorical elephants,
  131. 07:23we can realize a world that benefits everyone.
  132. 07:26So I implore all of you,
  133. 07:28take heed of the grass beneath your feet.
  134. 07:31Our collective future depends on it.
  135. 07:35Asanteni.
  136. 07:36(Cheers and applause)