Meus estudos..

Coisas que eu gostaria de fazer.. Vai no site de relação com investidores e pega a ultima apresentação X deles.

O que a IA não tem ? Feedback.. 70% da comunicação não verbal Capacidade de evoluir.. por exemplo, a minha evolução de atas de reunião.

Stuart Russell - Living With Artificial Intelligence The Biggest Event in Human History https://www.bbc.co.uk/sounds/play/m001216j

https://www.bbc.co.uk/sounds/brand/b00729d9

Ai - não substitui a decisão

Politano REFLEXÕES, pois ligamos isso diretamente a AI também:
1. The cost of cognition, the cost of creativity will massively decline with generative AI. How will that change productivity, innovation, etc.?
2. Everything on the internet in the world, good and bad and now can “generate”, it can predict the next word, the next pixel…
3. It is still in its infancy - apesar de já ter evoluído muito - mas tem MUITA pista ainda, por isso a afirmação… 🙂
4. Are machines going to replace humans? This is the question since the Industrial Revolution…
5. Lastly: Ironically, robots are building cars, so “robots” can drive those cars. End-to-end with almost no human intervention…
AI – Moving beyond Usage Policy, Pilot Use Cases and Strategy:
- Wilkery Marie Yeargan’s great share of AI Reference Architecture. It is strongly recommended that AI be integrated into existing EA and Governance Processes whenever possible vs creating a completely separate set of processes. This goes right along with the new research – Don’t rush to appoint a Chief AI Officer (see top of David’s new content below)
- Hot off the press last week - 9 Regulatory and Legal Implications of Generative AI - https://www.gartner.com/document/5308363
- Strongly suggest you watch Chris Howard’s series on AI Here is a sample - Generative AI for Real-World Business Productivity - https://www.gartner.com/document/4667099
  
  AI hype has reached a fever pitch, and investments are ramping up, sometimes without effective enterprise architecture guardrails. EA leaders should use this tool, Documenting AI Reference Architectures, to help their organization produce AI reference architectures that can be used to guide solution design and delivery teams.
  
  When to Use:
  - This tool is designed to help EA leaders scope, define, document and communicate AI reference architectures required to support business and IT outcomes. It provides clear guidance to teams of the architecture building blocks and approach to be used for solution architecture design and delivery.
  - The tool is a fully customizable PowerPoint template that demonstrates how to complete a full-scope AI reference architecture or minimum viable architecture (MVA) using a fictitious enterprise called QwaserZeeta Corp. The tool provides EA leaders guidance on completing the steps shown in Figure 1.
  Figure 1. 10 Steps to Build Your AI Reference Architecture
  
  If you or your clients have any questions about this research topic, or would like to speak with the Gartner analyst(s) Andreas Frangou, Shubhangi Jena, & Marcus Blosch who published this research please reach out to RES at +1 203-316-1200 or email at inquiry@gartner.com. For more guidance around similar topics please take a look at our key initiative(s): Enterprise Architecture Delivery & Artificial Intelligence
  
  Related research recommendations can also be found below:
  
  Recommended by the Author:

GenAI Introspective

ChatGPT Prompts

https://workflowy.com/template/chatgpt-prompts/

What Generative AI Means for Your Talent Strategy

Generative AI will have a dramatic impact on talent that is hard to predict. To anticipate and prepare for this disruption, executive leaders must factor AI into their workforce strategies, considering both future demand and how the enterprise will apply the technology.

Published: 03 Aug 2023

Our Investment in Anthropic: The Foundation Layer for Generative AI - Menlo Ventures (menlovc.com)

VM as GenAI PM

•Tipo de Pergunta: Processo… Qual o processo para

•Additional options i can awnser:

–Point

Study

Embeddings?

What are embeddings?

Read the guide from OpenAI
Literal: Embedding something turns it from image/text/audio into a list of numbers. 🖼️ or 📄 ⇒ [1.2, 2.1, ....]. This process makes documents “understandable” to a machine learning model.
By analogy: An embedding represents the essence of a document. This enables documents and queries with the same essence to be “near” each other and therefore easy to find.
Technical: An embedding is the latent-space position of a document at a layer of a deep neural network. For models trained specifically to embed data, this is the last layer.
A small example: If you search your photos for “famous bridge in San Francisco”. By embedding this query and comparing it to the embeddings of your photos and their metadata - it should return photos of the Golden Gate Bridge.

What are embeddings?

](https://platform.openai.com/docs/guides/embeddings/what-are-embeddings)

OpenAI’s text embeddings measure the relatedness of text strings. Embeddings are commonly used for:

Search (where results are ranked by relevance to a query string)
Clustering (where text strings are grouped by similarity)
Recommendations (where items with related text strings are recommended)
Anomaly detection (where outliers with little relatedness are identified)
Diversity measurement (where similarity distributions are analyzed)
Classification (where text strings are classified by their most similar label)

https://www.youtube.com/watch?v=Ue7M826xj5M

RAG

Tem emails do medium

Retrieval Augmented Generation(RAG) — Chatbot for database with LlamaIndex (Text2SQL) | by A B Vijay Kumar | Medium Retrieval Augmented Generation(RAG) — Chatbot for database with LlamaIndex… Implement the RAG technique using Langchain, and LlamaIndex for conversational chatbot on a database, using…

Good Example:

https://techstrong.ai/articles/an-ai-powered-document-search-app-with-chroma/

Worked fine:

“C:\dev\md\files\GenAI RAG Test - Geckos.docx”

Good video:

Retrieval Augmented Generation with OpenAI/GPT and Chroma - YouTube