65% of Sequoia Capital’s Investment Portfolio Companies Harness AI and LLMs
In Brief
Sequoia Capital’s survey revealed widespread adoption of AI and large language models (LLMs) in 33 companies, with OpenAI’s GPT and Anthropic’s GPT being the favorite foundation model APIs.
An enlightening survey conducted by Sequoia Capital has unveiled the widespread adoption of AI and large language models (LLMs) in various companies across their network. From small startups to large enterprises, 33 companies were interviewed to gain insights into their utilization of LLMs and the emerging stacks that accompany them.
- LLMs in Action: Language models have found their way into a multitude of products. Companies are leveraging LLMs to enhance code autocompletion, data science workflows, chatbot interactions, and a wide range of industries, including visual art, marketing, sales, legal, accounting, productivity, search, e-commerce, and travel planning. The applications are diverse and expanding.
- The New Stack: The emerging stack for LLM applications revolves around language model APIs, retrieval mechanisms, and orchestration. Open-source solutions are also gaining popularity.
Summary:
- Adoption Status: Around 65% of companies have integrated LLM applications into their production environment, while the remaining 35% are still in the experimentation phase.
- API Preferences: Among the foundation model APIs, OpenAI’s GPT stands out as the favourite, chosen by 91% of the surveyed companies. Anthropic’s interest has also grown to 15%. Some companies even utilize multiple models.
- Retrieval Mechanism: 88% of the companies consider a retrieval mechanism, such as a vector database, as a crucial part of their stack. This retrieval mechanism aids in providing context to the model, improving result quality, reducing inaccuracies, and addressing data freshness concerns.
- LLM Orchestration and Development Framework: 38% of the companies express interest in frameworks like LangChain for LLM orchestration and application development. Adoption has seen a recent increase.
- Monitoring and Testing Tools: Less than 10% of companies are actively seeking tools to monitor LLM outputs, cost, performance, and conduct A/B testing. However, interest in these areas is expected to rise as larger enterprises and regulated industries adopt LLMs.
- Complementary Generative Technologies: Some companies are exploring the combination of generative text and voice technologies, which presents an exciting area of growth.
- Custom Model Training: 15% of the companies have developed custom language models either from scratch or using open-source solutions. This trend has seen significant growth in recent months. These custom models require a separate stack comprising compute resources, model hubs, hosting platforms, training frameworks, and experiment tracking tools.
The Future Outlook:
- Unique Context: Companies aspire to customize language models to suit their specific needs and contexts.
- Blending Stacks: Although currently separate, the stack for LLM APIs and the custom model training stack are gradually merging over time.
- Trustworthiness: To ensure full adoption, language models must improve their output quality, prioritize data privacy and security concerns.
- Multi-Modal Applications: Language model applications will increasingly involve multiple modes of interaction, such as text and voice.
The adoption of AI language models is reshaping the way companies develop their products. The survey conducted by Sequoia Capital reveals the prevalent use of LLMs and the evolving stacks that support their implementation. As AI continues to progress rapidly, companies are customizing models, seeking trustworthiness, and exploring new frontiers.
Read more about AI:
Disclaimer
In line with the Trust Project guidelines, please note that the information provided on this page is not intended to be and should not be interpreted as legal, tax, investment, financial, or any other form of advice. It is important to only invest what you can afford to lose and to seek independent financial advice if you have any doubts. For further information, we suggest referring to the terms and conditions as well as the help and support pages provided by the issuer or advertiser. MetaversePost is committed to accurate, unbiased reporting, but market conditions are subject to change without notice.
About The Author
Damir is the team leader, product manager, and editor at Metaverse Post, covering topics such as AI/ML, AGI, LLMs, Metaverse, and Web3-related fields. His articles attract a massive audience of over a million users every month. He appears to be an expert with 10 years of experience in SEO and digital marketing. Damir has been mentioned in Mashable, Wired, Cointelegraph, The New Yorker, Inside.com, Entrepreneur, BeInCrypto, and other publications. He travels between the UAE, Turkey, Russia, and the CIS as a digital nomad. Damir earned a bachelor's degree in physics, which he believes has given him the critical thinking skills needed to be successful in the ever-changing landscape of the internet.
More articles
Damir is the team leader, product manager, and editor at Metaverse Post, covering topics such as AI/ML, AGI, LLMs, Metaverse, and Web3-related fields. His articles attract a massive audience of over a million users every month. He appears to be an expert with 10 years of experience in SEO and digital marketing. Damir has been mentioned in Mashable, Wired, Cointelegraph, The New Yorker, Inside.com, Entrepreneur, BeInCrypto, and other publications. He travels between the UAE, Turkey, Russia, and the CIS as a digital nomad. Damir earned a bachelor's degree in physics, which he believes has given him the critical thinking skills needed to be successful in the ever-changing landscape of the internet.
