Opening Keynote: The Next Era − We Shape AI, AI Shapes Us l Gartner IT Symposium/Xpo
16 Nov 2023 (10 months ago)
Intro
- The keynote is focused on AI and its impact on the relationship between humans and machines.
- AI is not just a technology or a business trend, but a profound shift in how humans and machines interact.
Story Time
- In a research workshop, participants developed an emotional attachment and refused to harm a baby dinosaur robot after interacting with it for an hour.
- Humans can form empathy with robots and develop positive relationships with them.
AI Shapes Us
- Machines have become more complex and intelligent, becoming more like humans and making human-machine interaction conversational.
- By 2030, 80% of people are predicted to interact with smart robots on a daily basis.
- CIOs have a key role in shaping AI and guiding organizations on how to derive the most value from it.
Our Kids
- Children's perception of technology is changing as they grow up in a world where machines talk back to them.
- Children may have unrealistic expectations of what machines can do, leading to potential issues when interacting with chatbots or AI systems.
Universalize Technology
- Machines are evolving from being our tools to becoming our consultants, protectors, coaches, friends, therapists, bosses, and customers.
- There are both positive and negative implications of the relationship between humans and machines, and it is important to be intentional about shaping this relationship.
AI Opportunity Radar
- AI comes in two flavors: everyday AI focused on productivity and gamechanging AI focused on creativity.
- Everyday AI enhances existing processes and can increase productivity by 5-20%.
- Gamechanging AI creates new types of value, products, services, and business models.
- Organizations should consider their opportunities in both everyday and gamechanging AI and determine which zones of the AI opportunity radar they want to focus on.
AI Back Office
- Everyday AI can supercharge the back office by automating tasks and liberating employees from mundane work.
- Examples include using AI to improve code quality or automating data analysis.
AI Front Office
- Gamechanging AI can supercharge the front office by enabling new capabilities and creating new results.
- Examples include using AI to detect wildfires or providing visual assistance to people with impairments.
Recap
- AI is more than just a technology and requires considering the human-machine relationship.
- CIOs need to guide the executive team in defining their AI ambition and populating the AI opportunity radar with everyday AI opportunities.
- There are two flavors of AI: everyday AI focused on productivity and gamechanging AI focused on creativity.
- Organizations should decide which zones of the AI opportunity radar they want to play in and involve the entire executive team in gamechanging AI initiatives.
How will AI affect your industry
- In the life sciences industry, AI can significantly accelerate drug discovery and development.
- Companies like Insilico Medicine have the capabilities to identify target diseases, generate new molecules, and predict clinical trial outcomes using AI.
- Gartner predicts that by 2025, more than 30% of new drugs and materials will be discovered using AI.
How will AI affect education
- AI can revolutionize education by providing personalized, interactive, and immersive learning experiences.
- Platforms like Khan Academy are using AI-powered teaching guides to enhance education.
- AI can enable virtual tutors, interactive learning experiences, and conversational interactions with historical figures.
Health warning
- Implementing game-changing AI comes with challenges and risks.
- Organizations need tolerance, executive patience, and financial resources to pursue AI initiatives.
- CFOs may be skeptical of AI investments, so organizations should expect scrutiny and show the ROI potential of AI.
Investment opportunities
- Organizations have three investment scenarios: defend, extend, and upend.
- Defending involves investing in quick wins and productivity assistants.
- Extending involves investing in custom applications and augmenting capabilities.
- Upending involves disruptive and expensive investments for potential high rewards.
AI ready principles
- CIOs need to establish AI ready principles to govern the use of AI.
- Principles define the guidelines and boundaries for human-to-machine relationships.
- Lighthouse principles based on values are essential to navigate the ethical challenges of AI.
You can't wait at a minimum
- Waiting for government regulations is not viable, as regulations often lag behind technology progress.
- Organizations need to make technology, economic, social, and ethical decisions when implementing AI.
- Lighthouse principles that align with values are crucial for ethical decision-making and navigating human-to-machine interactions.
Lighthouse principles are critical
- Lighthouse principles provide guidance for vendor selection and ethical AI relationships.
- Principles should be specific, unambiguous, and aligned with values.
- Principles help organizations make ethical and responsible decisions when interacting with machines.
AI ready data
- Only 4% of organizations have AI ready data.
- AI ready data is secure, enriched, fair, accurate, and governed by principles.
- Organizations should focus on making data that serves their AI ambitions AI ready, rather than trying to make all data AI ready.
Use Genai
- AI ready security is crucial to mitigate the risks and threats posed by AI.
- Bad actors can exploit AI models to steal private data.
- Organizations need to be aware of direct and indirect security threats associated with AI and implement measures to protect against them.
Counterfeit Reality
- Counterfeit reality is a situation where it's difficult to distinguish between what is real and what is fake.
- Indirect prompt injection is a scary concept where the prompt is modified after the user inputs it and before the model generates a response.
- Examples of counterfeit reality include the USSR's blue plague incident, which was entirely made up by Reddit users using generative AI.
- Generative AI can also be used to create damaging news stories about companies, spreading them over the internet through social media and numerous websites.
- Traditional security tools are not effective in solving the problem of counterfeit reality, and new tactics are needed to address it.