Tech is no issue. Challenges in AI are people-oriented - DSBoost #26

💬 Interview of the week

Enjoy our interview with Zeng:

• What did you study/are you studying (if your background is different from DS, how did you end up in the field)?

  I studied hotel management. Not AI related at all. I started to play with AI art last year around September and fell in love with it.

• What are your favorite resource sites and books (ML/AI)? 

  I love reading Rundown by Rowan Cheung and AI Supremacy by Michael Spencer.

• What got you starting WhoWhatWhyAi newsletter? 

  I got to know Brian Ball on Twitter, he is an amazing programmer who co-created AppZapper. We talk about AI development and he shares his insights. We found out that we can be a good team to write this newsletter together. 

• What do you enjoy the most in making AI art? 

  I see AI art as a therapy. Seeing words turn into a beautiful art is amazing. I remembered making my first AI art on NightCafe studio. That feeling of contentment and excitement was unforgettable. And I am so glad that with my skill, I started FemmeStock. It is a monthly subscription service for AI feminine stock images. I have $29 MRR at the moment. lol

• What tools do you use the most / favorite tools? 

  For AI art, I use LeonardoAI. And I use Aug X Lab for video generation, it is so useful. I turn my newsletter into videos with it. You can check out on WhoWhatWhyAi YouTube channel.  

• Do you use ChatGPT or other Al tools during your work? 

  Yes, I use ChatGPT. And I use GoCharlie AI for content repurposing. Turning newsletter into blogs and Tweets and making marketing copy.

• If so, how do they help you? Do they change your approach to problems?

  All these AI tools save me time. As a mother, I use chatGPT for work and for personal use such as making recipes and bedtime stories for kids. 

• What are you currently learning or improving (topics you are interested in nowadays)?

  I am exploring AI video prompt. I love how words can create art, and now create videos! I am excited.

• Can you share a fun fact about yourself?

  I use chopstick to eat potato chips and peanuts.

🎙️ Podcast of the week

SuperDataScience 698: How Firms Can Actually Adopt A.I., with Rehgan Avon

Here you can find the key takeaways that Rehgan Avon, Co-Founder & CEO of AlignAI, mentioned in the podcast about AI adoption by companies:

• Perception of AI has changed significantly over the years. In the past, industries like insurance and banking used simplistic techniques for auditability and explainability purposes. However, with the rise of generative models like ChatGPT, companies are now more interested in exploring the potential of AI and leveraging large language models (LLM).

• The promise of ROI (Return on Investment) from AI is more evident now than ever before. Companies are willing to invest in AI as the tools and resources for building and fine-tuning models have become more accessible and user-friendly.

• The barriers to entry for adopting AI are lower, with the availability of low-code and no-code options. This allows non-technical individuals to experiment and explore the possibilities of AI, leading to greater innovation within organizations.

• The challenges in AI adoption are primarily people-oriented rather than technical. Companies need to focus on educating their workforce about AI concepts, frameworks, tools, and their specific roles and responsibilities in the AI adoption process.

• AI adoption should be approached incrementally. Starting with simpler, more tangible use cases can help build confidence and demonstrate the power of AI within the organization.

• The education gap in AI adoption includes understanding concepts, frameworks, tools, roles, and responsibilities. Bridging this gap requires targeted education and training to align the workforce with AI initiatives.

🧵 Featured threads

🤖 What happened this week?

OpenAI Quietly Shuts Down Its AI Detection Tool.

In January, OpenAI introduced an AI detection tool called AI Classifier, designed to identify content generated using AI tools like ChatGPT. However, the tool has been shut down after six months due to its low accuracy. OpenAI stated that they are working on improving the technology and are researching better techniques for detecting AI-generated text, as well as planning to develop mechanisms to identify AI-generated audio and visual content. The rapid growth of sophisticated AI tools has led to the emergence of a cottage industry of AI detectors.

ChatGPT for Android launches next week

OpenAI is releasing the Android version of the app for ChatGPT next week after launching on iOS in May.