.
biotech
biotech
Biotechnology and Synthetic Biology
.
cryptography
crypto
Cryptography
.
materialscience
nano
Materials Science
.
neuroscience
neuroscience
Neuroscience
.
nuclear
nuclear
Nuclear Technologies
.
robotics
robotics
Robotics
.
semiconductor
semiconductors
Semiconductors
.
space
space
Space
.
Sustainable-Energy-Technologies_133px.jpg
energy
Sustainable Energy Technologies

Key Takeaways

•   AI is a foundational technology that is advancing other scientific fields and, like electricity and the internet, has the potential to transform how society operates.

•   Even the most advanced AI has many failure modes that are unpredictable, not widely appreciated, not easily fixed, not explainable, and capable of leading to unintended consequences.

•   There is substantial debate among AI experts about whether AI poses a long-term existential risk to humans, and whether the most important risks are also current weaknesses of AI.

Icons_card_AI.png

Overview

Artificial intelligence (AI) is the ability of computers to perform functions associated with the human brain, including perceiving, reasoning, learning, interacting, problem solving, and exercising creativity. AI promises to be a fundamental enabler of technological advancement and progress in many fields, arguably as important as electricity or the internet.

SUBFIELDS

AI has three core subfields; the boundaries between them are often fluid.

  • Computer vision (CV) enables machines to recognize and understand visual information, convert pictures and videos into data, and make decisions based on the results.

  • Machine learning (ML) enables computers to perform tasks without explicit instructions, often by generalizing from patterns in data. ML includes deep learning that relies on multilayered artificial neural networks to model and understand complex relationships within data. 

  • Natural language processing (NLP) equips machines with capabilities to understand, interpret, and produce spoken words and written texts.
INPUTS TO MACHINE LEARNING
 

Most of today’s AI is based on machine learning (ML), though it draws on other subfields. ML requires data and computing power, often on an enormous scale. Data can take various forms, including text, images, videos, sensor readings, and more. The quality and quantity of data play a crucial role in determining the performance and capabilities of AI models. Without sufficient high-quality data, AI models may generate inaccurate or biased outcomes. Furthermore, the hardware costs of training leading AI models are substantial. For example, reports have estimated that the training of GPT-4, ChatGPT’s more capable cousin, costs at least a few hundred million dollars. Currently, only a select number of large US companies have the resources to build cutting-edge models from scratch. 

REGULATION
 

Research on foundational AI technologies is difficult—if not impossible—to regulate, especially when other nations have strong incentives to carry on regardless of actions taken by US policymakers. The same applies to voluntary restrictions on research by companies concerned about competition. Regulation of specific applications of AI may be more easily implemented, in part because of existing regulatory frameworks in application domains such as health care, finance, and law.

 

Over the Horizon

AI OPPORTUNITIES   
 

AI users will not be limited to those with specialized training; instead, the average person will interact directly with sophisticated AI applications for a multitude of everyday activities. While AI can automate a wide range of tasks, it has promise in enabling people to do what they are best at doing. AI systems can work alongside people, complementing and assisting rather than replacing them. Key sectors poised to take advantage of AI include health care, agriculture, law, and the logistics and transportation field.

 

AI RISKS    

The primary challenge of bringing AI innovation into operation is risk management. Some of the known issues with today’s leading AI models include:

  • Explainability: Today’s AI is for the most part incapable of explaining how it arrives at a specific conclusion. Explanations are not always necessary, but in cases such as medical decision making, they may be critical.

  • Bias and fairness: Machine learning models are trained on existing datasets, which means that any bias in the data can skew results. (For example, using historical employment information at a particular firm to predict which job applicants are most desirable may lead to hiring preferences for men.)

  • Vulnerability to spoofing: For many AI models, data inputs can be tweaked to fool them into drawing false conclusions.

  • Deepfakes: AI provides the capability for generating highly realistic but entirely inauthentic audio and video, with concerning implications for courtroom evidence and political deception.

  • Overtrust: As trust in AI grows, the risk of overlooking errors, mishaps, and unforeseen incidents also grows.
  • Hallucinations: AI models can generate results or answers that seem plausible but are completely made up, incorrect, or both.

Report Preview: Artificial Intelligence

Faculty Council Advisor

fei-fei-li_profilephoto.jpg
Fei-Fei Li
Author
Fei-Fei Li

Fei-Fei Li is the Sequoia Professor of Computer Science and professor, by courtesy, of psychology at Stanford University. She serves as codirector of Stanford’s Human-Centered AI Institute and as an affiliated faculty at Stanford Bio-X. Her current research includes cognitively inspired AI, machine learning, computer vision, and ambient intelligent systems for health-care delivery. She received her PhD in electrical engineering from the California Institute of Technology.

View Bio
fei-fei-li_profilephoto.jpg
Fei-Fei Li

Fei-Fei Li is the Sequoia Professor of Computer Science and professor, by courtesy, of psychology at Stanford University. She serves as codirector of Stanford’s Human-Centered AI Institute and as an affiliated faculty at Stanford Bio-X. Her current research includes cognitively inspired AI, machine learning, computer vision, and ambient intelligent systems for health-care delivery. She received her PhD in electrical engineering from the California Institute of Technology.

Access the Complete Report

Read the complete report.

Explore

Date Range
CONTENT TYPE

Select Content Type

  • News
  • Article
  • Videos
  • Podcasts
  • Events
AUTHORS

Select Author

  • Condoleezza Rice
  • John Taylor
  • Jennifer Widom
  • Amy Zegart
  • Herbert Lin
  • Hon. Jerry McNerney
  • Hon. Robert Gates
  • Hon. Steven Chu
  • Hon. Susan M. Gordon
  • John Hennessy
  • Lloyd B. Minor
  • Mary Meeker
  • Peter Scher
  • Thomas M. Siebel
  • Zhenan Bao
  • Dan Boneh
  • Yi Cui
  • Simone D’Amico
  • Drew Endy
  • Siegfried Glenzer
  • Mark A. Horowitz
  • Fei-Fei Li
  • Allison Okamura
  • Kang Shen
  • Eric Schmidt
FOCUS AREAS

Artificial Intelligence

  • Artificial Intelligence
  • Biotechnology Synthetic Biology
  • Sustainable Energy Technologies
  • Cryptography
  • Materials Science
  • Neuroscience
  • Nuclear Technologies
  • Robotics
  • Semiconductors
  • Space
  • Technology Test Page
Date (field_date)
Read More
Space
Article
Books
A Deeper Dive into Space

Stanford University experts survey the aeronautics and astronautics research landscape on campus, covering topics including multisatellite autonomy; in-space logistics, servicing, assembly, and manufacturing (ISAM); space sustainability; spacecraft systems and…

June 17, 2024 by Herbert Lin, Simone D’Amico
Read More
Semiconductor Day at the Hoover Institution
News
Books
Experts, Diplomats, and Industry Representatives Discuss US Semiconductor Future after Passage of CHIPS Act

As dollars from the CHIPS and Science Act flow to strengthen the US semiconductor industry, senior leaders from the US government, industry, academia, and international partners met at the Hoover Institution to discuss current challenges and the field’s next…

June 13, 2024
Read More
SETR_CongressionalFellowship
News
Books
Stanford Emerging Technology Review Congressional Fellowship Program

The Hoover Institution and Stanford School of Engineering appreciate the important role of Congressional staff as they advise Members on the critically important policy issues that routinely come before Congress.

March 21, 2024
Read More
Space Innovation Roundtable
News
Books
Senior Leaders in Government, Industry, and Academia Convene for Discussions on Space Innovation and Commercial Integration
March 08, 2024
Read More
Artificial Intelligence
News
Books
US Wants Cloud Firms to Report Foreign Users Building AI

US Commerce Secretary Gina Raimondo said her department is exploring how to force cloud companies to disclose when a foreigner taps their computing power to fuel artificial intelligence applications, signaling the next phase of the tech war between Washington…

January 26, 2024
Read More
artificial intelligence
News
Books
Raimondo considers cloud reporting rules for foreign AI developers

Commerce Secretary Gina Raimondo said Friday that she is considering requiring cloud server providers to report details on whether foreign users are developing artificial intelligence models, an extension of the Biden administration’s efforts to institute AI…

January 29, 2024
Read More
artificial intelligence
News
Books
OpenAI and Other Tech Giants Will Have to Warn the US Government When They Start New AI Projects

The Biden administration is using the Defense Production Act to require companies to inform the Commerce Department when they start training high-powered AI algorithms.

January 26, 2024 by Condoleezza Rice, Fei-Fei Li
Read More
technologyiStock-1328282379
Videos
Books
Commerce Secretary and Others on AI and Innovation

Commerce Secretary Gina Raimondo, former Secretary of State Condoleezza Rice, and Stanford University computer scientist Fei-Fei Li discussed artificial intelligence and technological innovation at an event hosted by the Hoover Institution in Washington, DC.…

January 26, 2024 by Condoleezza Rice, Fei-Fei Li
Read More
SETR
News
Books
Stanford Emerging Technology Review Launches in Washington DC

Contributor scholars, Stanford engineers all, of the Stanford Emerging Technology Review brought their insights to the nation’s capital last week, launching the initiative with events in the nation’s capital.

January 29, 2024
Read More
technologyiStock-1328282379
Events
Books
A Conversation On The State Of Scientific And Technological Innovation

The U.S. Department of Commerce, together with the Hoover Institution and Stanford School of Engineering, invite you to a conversation on the state of scientific and technological innovation and the launch of Stanford Emerging Technology Review at JP Morgan…

January 26, 2024

You May Also Like

.
Artificial Intelligence
US Wants Cloud Firms to Report Foreign Users Building AI
.
artificial intelligence
Raimondo considers cloud reporting rules for foreign AI developers
.
artificial intelligence
OpenAI and Other Tech Giants Will Have to Warn the US Government When They Start New AI Projects
.
technologyiStock-1328282379
Commerce Secretary and Others on AI and Innovation
.
Artificial Intelligence
Stanford aims to help policy makers prepare for AI, robotics and more
.
Artificial Intelligence
Stanford launches emerging-tech project co-led by Hoover Institution’s Condoleezza Rice
.
Hoover research fellow Herbert Lin, the director and editor-in-chief of the Stanford Emerging Technology Review explains that advancements in a single field of emerging technology leads to advancements in others.
Hoover Institution and School of Engineering launch emerging technology review
.
AI Robot
Stanford AI professor Fei-Fei Li says we need more human-centered technology. Still, she had to convince herself to share her own story
.
Robot
AI is at an inflection point, Fei-Fei Li says
.
Artificial Intelligence
Fei-Fei Li Started an AI Revolution by Seeing Like an Algorithm
.
Artificial intelligence
Trailblazing computer scientist Fei-Fei Li on human-centered AI
.
Drone
Technology Applications By Policy Area
.
Globe
Cross-Cutting Themes
.
Stanford
Executive Summary
.
Binary
Foreword
overlay image