How to get the upside of AI without the downside Max Tegmark AI - - PowerPoint PPT Presentation

▶
how to get the upside of ai without the downside
SMART_READER_LITE
LIVE PREVIEW

How to get the upside of AI without the downside Max Tegmark AI - - PowerPoint PPT Presentation

Jun 27, 2023 252 likes •711 views

How to get the upside of AI without the downside Max Tegmark AI is getting more powerful Boston Dynamics: Atlas SpaceX NVIDIA AI can solve problems... How many lives can AI save? Over a million road deaths/year, Most soon preventable

slide-1
SLIDE 1

How to get the upside of AI 
 without the downside

Max Tegmark

slide-2
SLIDE 2
slide-3
SLIDE 3

AI is getting more powerful

slide-4
SLIDE 4 Boston Dynamics: Atlas
slide-5
SLIDE 5 SpaceX
slide-6
SLIDE 6 NVIDIA
slide-7
SLIDE 7
slide-8
SLIDE 8
slide-9
SLIDE 9

AI can solve problems...

slide-10
SLIDE 10

Over a million road deaths/year, most caused by human error Most soon preventable by AI?

How many lives can AI save?

slide-11
SLIDE 11

Bad hospital care kills more than twice as many as cars in US Most are preventable with AI

How many lives can AI save?

Source: Scientific American, http://tinyurl.com/medaccidents

slide-12
SLIDE 12

A recent Dutch study showed that AI diagnosis

  • f prostate cancer using MRI was as good as

that of human radiologists

AI diagnosis - prostate cancer

Source: Litjens et al 2015, Eur Radiol. 25, 3187–3199, http://tinyurl.com/prostate-ai

slide-13
SLIDE 13

A recent Stanford study showed that AI could diagnose lung cancer using microscope images even better than human pathologists

AI diagnosis - lung cancer

Source: Snyder et al 2016, http://tinyurl.com/lungcancer-ai

slide-14
SLIDE 14

A recent Google DeepMind study showed that AI could diagnose 50 retinal diseases from optical coherence tomography as human

  • ptalmologists

AI diagnosis - blindness

Source: De Fauw et al 2018, Nature Medicine, 24, 1342–1350

slide-15
SLIDE 15

AI diagnosis - ultrasound

  • NB. I have a conflict of

interest, since I'm advisor to the company making the Butterfly IQ

$20k-$75k $2k

Democratized by requiring 1) less money: 2) less training: AI guides non-experts to place

probe correctly & to analyze results

slide-16
SLIDE 16

Hemoglobin amino acid sequence: MVLSPADK...

AI for Protein Folding:

slide-17
SLIDE 17

43 mystery proteins 25 done best by AlphaFold AI 3 done best by #2 team of 98

CASP13 competition results, Dec. 2018:

AI for Protein Folding:

...and the AlphaFold team had almost no biology experience!

slide-18
SLIDE 18

Input amino acid sequence: TDELLERLRQLFEELHERGTEIVVEVHINGERDEIRVRNISKEELKKLLERIREKIEREGSSEVEVNVHSGGQTWTFNEK Protein shape predicted by AlphaFold:

slide-19
SLIDE 19
slide-20
SLIDE 20

...and AI can cause problems

slide-21
SLIDE 21
slide-22
SLIDE 22
slide-23
SLIDE 23
slide-24
SLIDE 24
slide-25
SLIDE 25
slide-26
SLIDE 26
slide-27
SLIDE 27
slide-28
SLIDE 28
slide-29
SLIDE 29
slide-30
SLIDE 30
slide-31
SLIDE 31

Let's draw the line!

slide-32
SLIDE 32

Acceptable Unacceptable

slide-33
SLIDE 33

Acceptable Unacceptable

Impossible?

slide-34
SLIDE 34

Acceptable Unacceptable

Impossible?

Biology is our role model!

  • Prof. Matthew Meselson

Nope!

slide-35
SLIDE 35

Bioweapons

Uses of biology:

Biology is our role model!

slide-36
SLIDE 36

Asilomar 1975: Asilomar 2017:

slide-37
SLIDE 37

Demis Hassabis, DeepMind, Founder & CEO Ilya Sutskever, OpenAI, Research Director, Co-Founder Yann LeCun, Director of AI Research at Facebook, Professor at New York University Yoshua Bengio, University of Montréal, head of the Montreal Institute for Learning Algorithms Stuart Russell, University of California, Berkeley, Professor of Computer Science, director of the Center for Intelligent Systems, co-author of the standard textbook Artificial Intelligence: a Modern Approach. Peter Norvig, Google, Research Director Ray Kurzweil, Google, Research Director; Inventor, Author and Futurist, author of The Singularity is Near and How to Create a Mind Jeff Dean, Google, Inc., Leader of Google Brain project, Google Senior Fellow, member of AAAI, ACM, National Academy of Engineering Tom Gruber, Apple; Co-Founder, CTO and VP Design for Siri Nils J. Nilsson, Past President of AAAI, Emeritus Kumagai Professor of Engineering, Stanford, member of AAAI Francesca Rossi, IBM & Padova, Professor of Computer Science, IJCAI President and Co-chair of AAAI committee on impact of AI and Ethical Issues, member of ACM Bart Selman, Cornell, Professor of Computer Science, co-chair of the AAAI presidential panel on long-term AI futures, member of ACM Guru Banavar, Vice President, Cognitive Computing, IBM Research Shane Legg, DeepMind, Co-Founder Mustafa Suleyman, DeepMind, Co-Founder Leslie Kaelbling, MIT, Professor of Computer Science and Engineering, member of AAAI, ACM, IEEE RAS …

Signatories of the 23 Asilomar AI Principles

slide-38
SLIDE 38

Demis Hassabis, DeepMind, Founder & CEO Ilya Sutskever, OpenAI, Research Director, Co-Founder Yann LeCun, Director of AI Research at Facebook, Professor at New York University Yoshua Bengio, University of Montréal, head of the Montreal Institute for Learning Algorithms Stuart Russell, University of California, Berkeley, Professor of Computer Science, director of the Center for Intelligent Systems, co-author of the standard textbook Artificial Intelligence: a Modern Approach. Peter Norvig, Google, Research Director Ray Kurzweil, Google, Research Director; Inventor, Author and Futurist, author of The Singularity is Near and How to Create a Mind Jeff Dean, Google, Inc., Leader of Google Brain project, Google Senior Fellow, member of AAAI, ACM, National Academy of Engineering Tom Gruber, Apple; Co-Founder, CTO and VP Design for Siri Nils J. Nilsson, Past President of AAAI, Emeritus Kumagai Professor of Engineering, Stanford, member of AAAI Francesca Rossi, IBM & Padova, Professor of Computer Science, IJCAI President and Co-chair of AAAI committee on impact of AI and Ethical Issues, member of ACM Bart Selman, Cornell, Professor of Computer Science, co-chair of the AAAI presidential panel on long-term AI futures, member of ACM Guru Banavar, Vice President, Cognitive Computing, IBM Research Shane Legg, DeepMind, Co-Founder Mustafa Suleyman, DeepMind, Co-Founder Leslie Kaelbling, MIT, Professor of Computer Science and Engineering, member of AAAI, ACM, IEEE RAS …

Signatories of the 23 Asilomar AI Principles

Endorsed by California Aug 30 2018

slide-39
SLIDE 39

Demis Hassabis, DeepMind, Founder & CEO Ilya Sutskever, OpenAI, Research Director, Co-Founder Yann LeCun, Director of AI Research at Facebook, Professor at New York University Yoshua Bengio, University of Montréal, head of the Montreal Institute for Learning Algorithms Stuart Russell, University of California, Berkeley, Professor of Computer Science, director of the Center for Intelligent Systems, co-author of the standard textbook Artificial Intelligence: a Modern Approach. Peter Norvig, Google, Research Director Ray Kurzweil, Google, Research Director; Inventor, Author and Futurist, author of The Singularity is Near and How to Create a Mind Jeff Dean, Google, Inc., Leader of Google Brain project, Google Senior Fellow, member of AAAI, ACM, National Academy of Engineering Tom Gruber, Apple; Co-Founder, CTO and VP Design for Siri Nils J. Nilsson, Past President of AAAI, Emeritus Kumagai Professor of Engineering, Stanford, member of AAAI Francesca Rossi, IBM & Padova, Professor of Computer Science, IJCAI President and Co-chair of AAAI committee on impact of AI and Ethical Issues, member of ACM Bart Selman, Cornell, Professor of Computer Science, co-chair of the AAAI presidential panel on long-term AI futures, member of ACM Guru Banavar, Vice President, Cognitive Computing, IBM Research Shane Legg, DeepMind, Co-Founder Mustafa Suleyman, DeepMind, Co-Founder Leslie Kaelbling, MIT, Professor of Computer Science and Engineering, member of AAAI, ACM, IEEE RAS …

Signatories of the 23 Asilomar AI Principles

Terah Lyons from PAI speaks this PM

slide-40
SLIDE 40

Acceptable Unacceptable

slide-41
SLIDE 41

Let’s build AI that

  • verpowers us!

empowers

slide-42
SLIDE 42

Let’s build AI that

  • verpowers us!

empowers

slide-43
SLIDE 43
slide-44
SLIDE 44