Janna Hart’s Post

Slow Adoption of Generative AI in Business The Wall Street Journal published an article last week titled "Generative AI Isn't Ubiquitous in the Business World—at Least Not Yet" (https://lnkd.in/gTbRiVTk). This article sparked my curiosity about the reasons behind the slow adoption of generative AI (Gen AI) in businesses.  After talking to a cross-section of customers over the past year, I've identified several key factors hindering widespread adoption:   Cost of AI: While consumer-facing AI tools may seem cheap or even free, they can be expensive for organizations. Take ChatGPT4, for example, which charges $10 per 1 million input tokens and $30 per 1 million output tokens. (Think of tokens as pieces of words; 1,000 tokens are roughly equivalent to 750 words.) For even a simple use case, this can translate to tens of thousands of dollars per month.   Lack of Talent: Implementing Gen AI effectively requires a range of skilled and trained professionals. Expertise is needed in Gen AI architecture, implementation, prompt engineering, content moderation, and governance.   Privacy and Security Concerns: Companies worry about their organizational information being fed into AI models. They also have concerns about employees inadvertently sending sensitive data outside firewalls through poorly defined prompts.   Accuracy and Hallucinations: Customers have encountered (and heard stories of) comical and sometimes alarming instances of AI generating nonsensical or misleading responses. This hesitancy to implement customer-facing applications without robust governance is understandable. Co-piloting with human intervention can further increase costs and decrease ROI (return on investment).  Solution:  Successful Gen AI implementations require an upfront AI strategy that addresses these concerns. This strategy should ensure sustainable economic value driven by responsible and ethical AI practices. Prevsiant has developed an AI Strategy approach that can help organizations achieve successful and cost-effective Gen AI implementations. You can review it here: https://lnkd.in/gcGN28sg

Everyone talks about AI. In fact, we also use #AI extensively in #VTDocs. But before delving into the nuances of AI and related technologies, it's important to understand the business context and the challenges it addresses. Let’s consider an example: When reviewing multiple documents, business analysts, project managers, and other professionals typically examine them for essential requirements: They identify dependencies within a document—perhaps paragraph three relates to paragraph 52 on page 19.  These interdependencies occur both within individual documents and across multiple ones.  Examples include standard operating procedures with specific policies, terms and conditions, or supplier agreements.  How does their process look like today? The analyst opens a document and skims through it from start to finish, noting any critical points or elements of interest.  They use the search function (Ctrl + F) to scan for terms like "warranty" or "indemnifications" in the contract. This method is time-consuming and ad-hoc. While experienced professionals might navigate quickly, those less experienced or new to the business need much more time and might overlook crucial details. VT Docs and its AI models offer a different approach: When uploading multiple documents into VT Docs, users instantly see common themes across documents – powered by AI.  Imagine a matrix where each column represents a document, and on the left, thematic frequencies are highlighted.  This visualization allows users to spot patterns quickly. For instance, if references to "indemnification" or variations of that appear across several documents, then VT Docs shows them automatically and provides an aggregated view. VT Docs utilizes an AI model specially designed for business and technical communications.  So, there's no "hallucination" or inaccuracy— the AI doesn't use a generative model (like ChatGPT) but relies on a model that can trace back to the precise text, ensuring 100% accuracy. This precision is non-negotiable and pivotal. Another consideration, especially concerning generative AI, is security and data integrity.  VT Docs and its AI model run completely isolated in the most secure locked-down environments. This ensures no data leaks or unauthorized learning from your data. AI has the potential to transform industries, but only if it maintains reliability and security and seamlessly integrates into existing workflows. At @VisibleThread, we aim to pioneer solutions that embody these principles, ensuring not just innovation but secure and dependable results. #documentreview #artificialintelligence #enterprise

Everyone talks about AI. In fact, we also use #AI extensively in #VTDocs. But before delving into the nuances of AI and related technologies, it's important to understand the business context and the challenges it addresses. Let’s consider an example: When reviewing multiple documents, business analysts, project managers, and other professionals typically examine them for essential requirements: They identify dependencies within a document—perhaps paragraph three relates to paragraph 52 on page 19.  These interdependencies occur both within individual documents and across multiple ones.  Examples include standard operating procedures with specific policies, terms and conditions, or supplier agreements.  How does their process look like today? The analyst opens a document and skims through it from start to finish, noting any critical points or elements of interest.  They use the search function (Ctrl + F) to scan for terms like "warranty" or "indemnifications" in the contract. This method is time-consuming and ad-hoc. While experienced professionals might navigate quickly, those less experienced or new to the business need much more time and might overlook crucial details. VT Docs and its AI models offer a different approach: When uploading multiple documents into VT Docs, users instantly see common themes across documents – powered by AI.  Imagine a matrix where each column represents a document, and on the left, thematic frequencies are highlighted.  This visualization allows users to spot patterns quickly. For instance, if references to "indemnification" or variations of that appear across several documents, then VT Docs shows them automatically and provides an aggregated view. VT Docs utilizes an AI model specially designed for business and technical communications.  So, there's no "hallucination" or inaccuracy— the AI doesn't use a generative model (like ChatGPT) but relies on a model that can trace back to the precise text, ensuring 100% accuracy. This precision is non-negotiable and pivotal. Another consideration, especially concerning generative AI, is security and data integrity.  VT Docs and its AI model run completely isolated in the most secure locked-down environments. This ensures no data leaks or unauthorized learning from your data. AI has the potential to transform industries, but only if it maintains reliability and security and seamlessly integrates into existing workflows. At @VisibleThread, we aim to pioneer solutions that embody these principles, ensuring not just innovation but secure and dependable results. #documentreview #artificialintelligence #enterprise

Everyone talks about AI. In fact, we also use #AI extensively in #VTDocs. But before delving into the nuances of AI and related technologies, it's important to understand the business context and the challenges it addresses. Let’s consider an example: When reviewing multiple documents, business analysts, project managers, and other professionals typically examine them for essential requirements: They identify dependencies within a document—perhaps paragraph three relates to paragraph 52 on page 19.  These interdependencies occur both within individual documents and across multiple ones.  Examples include standard operating procedures with specific policies, terms and conditions, or supplier agreements.  How does their process look like today? The analyst opens a document and skims through it from start to finish, noting any critical points or elements of interest.  They use the search function (Ctrl + F) to scan for terms like "warranty" or "indemnifications" in the contract. This method is time-consuming and ad-hoc. While experienced professionals might navigate quickly, those less experienced or new to the business need much more time and might overlook crucial details. VT Docs and its AI models offer a different approach: When uploading multiple documents into VT Docs, users instantly see common themes across documents – powered by AI.  Imagine a matrix where each column represents a document, and on the left, thematic frequencies are highlighted.  This visualization allows users to spot patterns quickly. For instance, if references to "indemnification" or variations of that appear across several documents, then VT Docs shows them automatically and provides an aggregated view. VT Docs utilizes an AI model specially designed for business and technical communications.  So, there's no "hallucination" or inaccuracy— the AI doesn't use a generative model (like ChatGPT) but relies on a model that can trace back to the precise text, ensuring 100% accuracy. This precision is non-negotiable and pivotal. Another consideration, especially concerning generative AI, is security and data integrity.  VT Docs and its AI model run completely isolated in the most secure locked-down environments. This ensures no data leaks or unauthorized learning from your data. AI has the potential to transform industries, but only if it maintains reliability and security and seamlessly integrates into existing workflows. At @VisibleThread, we aim to pioneer solutions that embody these principles, ensuring not just innovation but secure and dependable results. #documentreview #artificialintelligence #enterprise

Everyone talks about AI. In fact, we also use #AI extensively in #VTDocs. But before delving into the nuances of AI and related technologies, it's important to understand the business context and the challenges it addresses. Let’s consider an example: When reviewing multiple documents, business analysts, project managers, and other professionals typically examine them for essential requirements: They identify dependencies within a document—perhaps paragraph three relates to paragraph 52 on page 19. These interdependencies occur both within individual documents and across multiple ones. Examples include standard operating procedures with specific policies, terms and conditions, or supplier agreements. How does their process look like today? The analyst opens a document and skims through it from start to finish, noting any critical points or elements of interest. They use the search function (Ctrl + F) to scan for terms like "warranty" or "indemnifications" in the contract. This method is time-consuming and ad-hoc. While experienced professionals might navigate quickly, those less experienced or new to the business need much more time and might overlook crucial details. VT Docs and its AI models offer a different approach: When uploading multiple documents into VT Docs, users instantly see common themes across documents – powered by AI. Imagine a matrix where each column represents a document, and on the left, thematic frequencies are highlighted. This visualization allows users to spot patterns quickly. For instance, if references to "indemnification" or variations of that appear across several documents, then VT Docs shows them automatically and provides an aggregated view. VT Docs utilizes an AI model specially designed for business and technical communications. So, there's no "hallucination" or inaccuracy— the AI doesn't use a generative model (like ChatGPT) but relies on a model that can trace back to the precise text, ensuring 100% accuracy. This precision is non-negotiable and pivotal. Another consideration, especially concerning generative AI, is security and data integrity. VT Docs and its AI model run completely isolated in the most secure locked-down environments. This ensures no data leaks or unauthorized learning from your data. AI has the potential to transform industries, but only if it maintains reliability and security and seamlessly integrates into existing workflows. At VisibleThread, we aim to pioneer solutions that embody these principles, ensuring not just innovation but secure and dependable results. #documentreview #artificialintelligence #enterprise

AI is the technology of the decade, but, as this Axios article points out, we still don't really know what to do with it. I see a lot of business leaders trying to implement AI in order to stay ahead of the competition, instead of implementing a solution to a problem. While I'll always advocate for exploring new technology, I don't recommend starting a multi-million dollar program for the sake of saying "we use AI." Instead of focusing on the capabilities of a ChatGPT or DALL-E, I'm significantly more interested in finding ways to leverage AI and machine learning (ML) for clear business value. One area that I am seeing business value demonstrated is in the area of decision intelligence. I see a lot of potential in, for example, the manufacturing space where an AI model can inform resource management by recommending what items should be produced in which factory or what quantity of products should be kept in certain warehouses to best meet consumer demand. In this use case, AI is able to account for more variables than a human could, and in doing so is able to develop more nuanced solutions. This type of decision intelligence is one area where I see the value of AI/ML starting to come to fruition and one of my predictions for how the technology will be used successfully in the coming years. Last week, I attended the Gartner Supply Chain Symposium and participated in several conversations with other business leaders about the practical application of AI, particularly when it comes to optimizing supply chains. From healthcare to utilities to manufacturing and CPG, there are a lot of potential benefits to adopting decision intelligence into existing supply chain processes (as well as other processes). I’m looking forward to continuing these conversations with our Sendero industry teams in the coming weeks.

A tool is only as good as the solution it provides!

"#AI #hallucinations – how can you help us with them?" If you're familiar with AI, you’re likely aware of the myriad risks associated with LLMs. One of those risks is AI hallucinations. AI hallucinations refer to situations where AI-generated outputs deviate from accurate representations of reality. → For instance, it's when you ask ChatGPT a question and the response is incorrect or misleading. One of the ways to counteract such risks is to deploy LLMs in areas where the likelihood of AI hallucinations is minimal. → For instance, leveraging LLMs to support search engine or chatbot mechanisms but not to paraphrase information. Instead of providing paraphrased answers prone to errors, the system retrieves and presents exact text fragments from relevant documents or sources. This method ensures reliability and minimizes the potential for AI hallucinations to mislead users. That's exactly the approach we suggest when implementing AI in organizations – to support processes rather than serve as the primary interface. If you do that too, you’ll mitigate many risks associated with AI (hallucinations being a perfect example). :::::::::::::::::::::::::: Follow me for more tips on implementing AI in your business in a way that makes the most sense.

Feel like generative AI has already taken over the world? Don't let the hype fool you. The real data might surprise you a little - both for Generative AI, and for AI in general. The reality is that many companies are recognizing the risks posed and challenges faced by incorporating these new systems ... and they're taking their time about it. Us too! Read for the full story . #ailaw #aigovernance #generativeai #responsibleai

Let's delve into the world of AI! Quality data, especially proprietary data, is the new gold when it comes to AI. What you do with that data is key – whether it's through proprietary modeling or training sets, processing non-monetizable data into a monetizable output can enhance various business functions. Speed is crucial too, but it's less important if you're working with poor data or inadequate modeling. A winning combination comprises quality data, proprietary modeling or training sets, and speed. Mastering these can lead a company to success and grab attention. On the flip side, using publicly available commercial data like ChatGPT, with subpar modeling or training, may not qualify a company as an AI player deserving an AI valuation premium. Distinguishing between authentic AI companies and those without essential qualities is crucial in this evolving landscape. Exploring the realm of AI companies reveals a mix of exciting and unique ventures, alongside those that may not meet the criteria to be termed as genuine AI companies. Exciting times lie ahead as we navigate through this dynamic industry!