Hi Everyone,

We always like you to get the most value out of the newsletter, in addition to the email version of the newsletter there is a supporting website where you can log on and get to all the previous versions. You can also, share, and comment and do all the usual things.

Also… we now are in audio!

Thanks for reading.

Paul

Table of Contents

• Introducing Web Search in ChatGPT

• Claude: Advancements in Data Analysis and Visualis …

• Google's Embedded Search

• AI Innovations in Creative Platforms

• Understanding AI Agents: Integration of Business P …

  • Debating AI Autonomy - Prioritising Human and AI D …

• Exploring Google NotebookLM

• Implementing AI in SMEs

• Establishing AI Decision-Making: A Five-Step Model

Introducing Web Search in ChatGPT

OpenAI has recently enhanced ChatGPT by integrating a web search feature, enabling the AI to access real-time information from the internet. This development allows ChatGPT to provide up-to-date responses, particularly beneficial for queries about current events, stock prices, or recent developments. The search functionality is powered by Microsoft's Bing, ensuring a broad and reliable data source.

However, users have observed instances where ChatGPT returns different answers for identical queries, a phenomenon known as "hallucination." This raises concerns about the consistency and reliability of AI-generated information. In traditional search engines, discrepancies can occur due to varying algorithms and data sources, but AI models like ChatGPT may amplify these inconsistencies. For business professionals, it's crucial to cross-verify AI-generated information with trusted sources to ensure accuracy.

It is going to be interesting to see how this affects performance marketing and SEO. In the immediate term, considering that bing has achieved 4% of the search market, outside the circle of ‘AI’ I wonder how many actual users will be aware of this.

The AI overviews from Google which started off extremely ‘variable’, actually now seem to be showing up in results in a fairly integrated and even un-noticeable way. It will be interesting to see how this impacts

• the ad revenue model

• ‘AI content is bad for SEO (according to performance marketing agencies)’… when Google themselves are generating the content using AI, and have been stoic on its impact on SEO…

Claude: Advancements in Data Analysis and Visualisation

Anthropic has introduced significant updates to its AI model, Claude, enhancing its capabilities in data analysis and visualisation. The new "Analysis Tool" enables Claude to write and execute JavaScript code, allowing for complex data processing and real-time insights. This feature transforms Claude into a versatile data analyst, capable of handling tasks such as cleaning, exploring, and analysing datasets systematically

Additionally, Claude now supports visual PDF analysis, enabling the AI to process documents containing images, charts, and graphs. This functionality is particularly beneficial for professionals dealing with comprehensive reports, as Claude can extract and interpret visual data, providing concise summaries and insights.

Claude is probably now my go to AI for general usage, including code and content support

Google's Embedded Search

Google has launched embedded search APIs, allowing developers to integrate Google's search capabilities directly into their applications. This integration enables businesses to offer users a seamless search experience within their platforms, leveraging Google's extensive search algorithms and data. The cost structure for these APIs varies based on usage and specific requirements, making it essential for businesses to assess their needs and budget accordingly.

For enterprises, embedding Google's search functionality can enhance user engagement and satisfaction by providing accurate and relevant information swiftly. However, it's important to consider the associated costs and ensure that the integration aligns with the organisation's strategic objectives.

However, the cost is $35 per 1000 queries, which is either to control demand or a weird price point, considering you can almost achieve the same results for 1000th of that cost using available alternatives.

Google have such a powerful toolset, but unlike all the other vendors, still seem to be struggling to find the routes to market and how to resonate with enterprise solutions teams, other than in deep AI research.

If you have logged in (or are a subscriber..) even more to read follows.