Noam M Shazeer

This inventor holds 83 USPTO granted patents and 22 published patent applications, plus 4 CIPO patents and 6 EPO patents, primarily in Neural Networks (CPC class G06N3-08). Top assignee: Google Inc.. Active years: 2005-2026. Education: Duke University (USA).

Graduated from:

Duke University (USA)

USPTO Granted Patents = 83 (Patent Applications = 22)

% Patents Active = 89.2

CIPO Patents = 4

EPO Patents = 6 (Patent Applications = 20)

Average Co-Inventor Count = 3.1

ph-index = 14

Forward Citations = 924(Granted Patents)

Forward Citations (Not Self Cited) = 783(Dec 10, 2025)

Inventors with similar research interests:

🔬 See more inventors

Location History:

Mountain View, CA (US) (2007 - 2016)
Stanford, CA (US) (2007 - 2018)
Palo Alto, CA (US) (2005 - 2024)

Company Filing History:

83 out of 33,920

Google Inc.

patents

Years Active: 2005-2026

Loading Chart...

Areas of Expertise:

Attention Mechanisms

Language Models

Image Generation

Sequence Transduction

Speech Recognition

Granular Architecture Search

Multi-Task Learning

Tensor Computations

Machine Learning

Auto-Regressive Models

Mixture of Experts

83 patents (USPTO):Sort by Forward CitationsExplore Patents

1. 12608616 - Attention neural networks with linear units (2026-04-21),

2. 12481834 - Machine-learned language models which generate intermediate textual analysis in service of contextual text generation (2025-11-25),

3. 12430515 - Machine-learned language models which generate intermediate textual analysis in service of contextual text generation (2025-09-30),

4. 12393840 - Granular neural network architecture search over low-level primitives (2025-08-19),

5. 12373688 - Granular neural network architecture search over low-level primitives (2025-07-29),

6. 12354005 - Attention-based decoder-only sequence transduction neural networks (2025-07-08),

7. 12353991 - Fast decoding in sequence models using discrete latent variables (2025-07-08),

8. 12299573 - Attention-based decoder-only sequence transduction neural networks (2025-05-13),

9. 12299572 - Attention-based decoder-only sequence transduction neural networks (2025-05-13),

10. 12271817 - Attention-based decoder-only sequence transduction neural networks (2025-04-08),

11. 12265903 - Distributing tensor computations across computing devices (2025-04-01),

12. 12254411 - Attention neural networks with linear units (2025-03-18),

13. 12217173 - Attention-based sequence transduction neural networks (2025-02-04),

14. 12148421 - Using large language model(s) in generating automated assistant response(s (2024-11-19),

15. 12142034 - Attention-based image generation neural networks (2024-11-12),

Noam Shazeer is a prominent figure in the field of artificial intelligence, widely recognized for his contributions to transformative AI technologies. A longtime Google engineer, Shazeer played a pivotal role in the AI boom, co-authoring the seminal 2017 paper 'Attention is All You Need,' which laid the foundation for modern generative AI systems.

After joining Google in 2000 as one of its early employees, Shazeer worked on several groundbreaking projects, including improving the search engine's spelling correction and developing AI systems designed to achieve human-level intelligence.

In 2021, after Google declined to release a chatbot he had developed, Shazeer left to co-found Character.AI with fellow Google engineer Daniel De Freitas. Character.AI gained significant attention, raising $150 million in 2022 and being valued at $1 billion, but faced challenges in scaling its business model. The startup offered users the opportunity to interact with chatbots simulating practical advice and even personalities like Elon Musk.

Despite these hurdles, Shazeer’s talent remained in high demand. In a major deal, Google agreed to pay around $2.7 billion to license Character. AI's technology, with Shazeer returning to Google as part of the arrangement. He now serves as Vice President at Google, playing a leading role in the development of its next-generation AI platform, Gemini.

Throughout his career, Shazeer has maintained a vision of AI as a tool with the potential to significantly impact society, from addressing loneliness to solving complex knowledge problems. Despite stepping away from his startup, he remains at the forefront of AI innovation, influencing the future trajectory of one of the most powerful tech companies in the world.

(Adapted from an article in WSJ)

Google’s $2.7 Billion Move to Re-Hire AI Genius Draws Attention