​AI Tech Circle​

Lost in AI use cases, remember to prioritize your AI Strategy and align it with the business.

Stay Ahead in AI with Weekly AI Roundup; read and listen on AITechCircle: May 25, 2024

Welcome to the weekly AI Newsletter, where I provide actionable ideas and tips to assist you in your job and business.

Before we start, share this week's updates with a friend or a colleague:

Today at a Glance:

• Prioritizing AI strategy aligned with your business outcomes
• AI Weekly news and updates
• Generative AI Usecase: Enhancing Employee Safety with Personalized OHS Training Using Generative AI
• Favorite Tip Of The Week & Potential of AI
• Things to Know: Frontier Safety Framework
• Open Tech Talk Podcast, the latest episode on Building a Successful AI SaaS Product
• AI/ML/GenAI courses to attend
• AI Events
• Tech Tools to try out
• Datasets for your next project
• Other Tech news

It is important to reduce barriers to adopting responsible AI use and challenges related to IT infrastructure, data, cybersecurity, and the workforce, particularly risks of Generative AI.

You need an organizational AI Strategy that will become the blueprint for your organization to start working on it. If you are lost in AI use cases, don't forget to put your AI strategy first and align it with your business.

Develop an enterprise AI strategy to Advance Responsible AI Innovation

Creating and linking your AI strategy to the organization’s strategic objective and navigating AI investments will help your organization achieve and adopt AI, driven not only by the use cases but also by the business objectives.

Based on the organization's maturity level, a very small percentage of organizations have extensive experience implementing and running AI solutions. The rest are entering a new venture that is exciting and rewarding. Therefore, it is essential to have the fundamentals set.

​

Below is the slide deck I have created to kick-start this area. These are my thoughts; feel free to share your view and update/edit as per your organizational goals.

The newly formed Enterprise AI Strategy must be designed to solve large-scale real-world problems. To create a successful Enterprise AI Strategy, it is first necessary to assess the skills and capabilities of your existing organization.

This includes an assessment of existing executive leadership to determine if a suitable AI Champion exists inside the organization or if someone needs to be brought in from the outside.

​

Embracing AI-Driven Business Transformation can revolutionize your organization’s efficiency and innovation. By aligning AI initiatives with your strategic goals, you can unlock new revenue streams, optimize operations, and gain a competitive edge.

An Intelligent Data Ecosystem is essential for maximizing AI's impact on your organization. You ensure accurate insights and better decision-making by organizing and managing your data effectively. Act now to streamline your data architecture and invest in scalable cloud solutions.

Investing in AI Talent Acceleration and fostering an innovation-driven culture is vital for your organization's success. You can quickly develop the skills needed for advanced AI projects by providing training and building collaborative teams. Start today to attract top talent and create a supportive environment that drives AI innovation and growth.

Ensuring AI Ethics and Risk Mitigation is crucial for your organization's responsible AI adoption. Ethical guidelines and risk management practices will protect your business and build stakeholder trust. Act immediately to establish a framework that ensures compliance and minimizes risks, securing your AI initiatives for the future.

You can download the complete slide deck from AITechCircle. It also contains the implementation roadmap covering the short-term, medium-term, and long-term plans.

​

In the coming weeks, we will cover each pillar in depth, considering one model/example organization. The ultimate objective is to prepare the AI strategy for this organization.

Weekly News & Updates...

The last two week's AI breakthroughs mark another leap forward in the tech revolution.

1. During Google I/O, Gemini 1.5 Flash, a lighter-weight model, was introduced. Gemini 1.5 Pro and 1.5 Flash are available in public preview with a 1 million token context window and a 2 million token window. Veo generates high-quality 1080p resolution videos in various cinematic and visual styles.
2. ​GPT-4o was announced; the “o” is for Omni. The model accepts and generates text, images, audio, and video. It supports 50+ languages, It also added voice Mode with new audio and video capabilities.
3. ​Falcon 2 is realeased by the UAE TIU, it is Open-Source, Multilingual, and Multimodal. Falcon 2 has 11B LLM trained on 5.5 trillion tokens with 11 billion parameters and 11B VLM has vision-to-language model (VLM) capabilities.
4. IBM has released a family of LLM Granite code models as open-source. It is a series of decoder-only Granite code models for code generative tasks, trained with code written in 116 programming languages. The models range in size from 3 to 34 billion parameters.
5. Cohere has launched Aya 23, a multilingual 8B and 35B open weights release. It covers language modeling capabilities for nearly half of the world's population. Aya-23-35B achieves the highest results for the languages covered in discriminative and generative multilingual benchmarks. Go through the technical report over here and download it from here​
6. ​News Corp and OpenAI have entered into a significant multi-year agreement, allowing OpenAI to use News Corp's news content. This partnership enables OpenAI to display information from News Corp’s publications in response to user inquiries and enhance its products. The goal is to provide users with reliable information, helping them make informed decisions based on trustworthy news sources.
7. ​LearnLM-Tutor, soon we will have access to the LLM fine-tuned for learning based on Gemini. This will be rolled out in Google Search, YouTube, and Android devices to learn complex topics with the support of Gen AI.

The Cloud: the backbone of the AI revolution

• ​Oracle Code Assist: AI companion to boost developer velocity. It is powered by large language models (LLMs) on OCI and fine-tuned and optimized for Java, SQL, and application development on OCI, Oracle Code Assist provides developers with context-specific suggestions.
• ​Every Company to Be an ‘Intelligence Manufacturer,’ Declares NVIDIA CEO Jensen Huang at Dell Technologies World.

Gen AI Use Case of the Week:

Generative AI use cases in the supply chain industry:

Enhancing Employee Safety with Personalized OHS Training Using Generative AI.

Areas covered are:

1. Business Challenges
2. AI Solution Description
3. Expected Impact/Business Outcome
4. Required Data Sources
5. Strategic Fit and Impact Rating
6. Practical Example of Implementation

Favorite Tip Of The Week:

Here's my favorite resource of the week.

• Do you want to start building with Meta Llama models? Step-by-step tutorials to help you get started with Llama 3.

Potential of AI

• ​Watch The Journey of Aya - Accelerating Multilingual AI Through Open Science. How 3,000 collaborators from 119 countries were brought together to work on Aya.

Things to Know

​Frontier Safety Framework from Google Deepmind is a set of protocols for proactively identifying future AI capabilities that could cause severe harm and implementing mechanisms to detect and mitigate them.

This framework follows the approach published in the policy paper by the UK Department for Science, Innovation & Technology, namely 'Emerging processes for frontier AI safety'​

1. Identifying Harmful Capabilities: Research was done on how a model could cause severe harm in high-risk areas, and the minimal capabilities needed for such harm were determined, known as “Critical Capability Levels” (CCLs). These CCLs guide the evaluation and mitigation strategies.
2. Periodic Model Evaluation: Regularly assess the advanced models to detect when they reach CCLs. We develop “early warning evaluations” to alert us when a model is nearing a CCL and run these evaluations frequently to provide timely notice.
3. Mitigation Plan Implementation: We implement a mitigation plan when a model passes early warning evaluations. This plan considers the balance of benefits and risks and focuses on security (preventing model exfiltration) and deployment (preventing misuse of critical capabilities).

The Opportunity...

Podcast:

• This week's Open Tech Talks episode 135 is "Building a Successful AI SaaS Product with Denzil Eden." She is the founder of Smarty ai

Apple | Spotify | Google Podcast | Youtube

Courses to attend:

• ​Learn Generative AI by building a project from DeapLearning. This provides a detailed project description, a step-by-step outline of the RAG process, and courses for you to learn.
• Learn how to implement Llama 3 From Scratch.
• ​Introduction to Data-Centric AI from MIT. This class covers algorithms to find and fix common issues in ML data and to construct better datasets, concentrating on data used in supervised learning tasks like classification.
• ​Natural Language Understanding from Stanford Online. Taught by professor Christopher Potts. Covers theoretical concepts from linguistics, natural language processing, and machine learning.

Events:

• ​AI World Congress 2024, May 30-31, London, UK
• ​The AI Retreat 2024, June 13, Dubai, UAE
• ​GITEX GLOBAL, Oct 14-18, 2024, Dubai, UAE

Tech and Tools...

• ​Khoj is an application that creates always-available, personal AI agents for you to extend your capabilities.
• ​MiniCPM-V is end-side multimodal LLMs (MLLMs) designed for vision-language understanding. Models take image and text as inputs and provide high-quality text outputs.

Data Sets...

• ​The 20 Newsgroups data set is a collection of approximately 20,000 newsgroup documents, partitioned (nearly) evenly across 20 different newsgroups.
• ​Legal Case Reports: A textual corpus of 4000 legal cases for automatic summarization and citation analysis.

Other Technology News

Want to stay on the cutting edge?

Here's what else is happening in Information Technology you should know about:

• Elon Musk predicts jobs will become ‘kinda like a hobby’: ‘The AI and robots will provide any goods and services you want’ as reported by Fortune.
• ​Microsoft introduced new Windows PCs designed for AI, Copilot+ PCs.

AI First Community to Learn & Share...

Have a question or need some assistance with your AI project, or maybe you want to be part of the thriving community to learn AI together,

​Click here to join the AI Tech Circle - It's your Community on Discord​

Download 100+ Gen AI use cases

That's it!

As always, thanks for reading.

Hit reply and let me know what you found most helpful this week - I'd love to hear from you!

Until next week,

Kashif Manzoor