The Daily Wild: 🎙️ Podcast

Pamela Gupta, CEO, Co-President at Trusted AI™: "Why we need to pay more attention to AI"

Yael Rozencwajg — Sun, 30 Mar 2025 00:01:14 GMT

New season, new episodes, new guests.

The AI revolution demands a new level of strategic foresight.

This season, we're exploring decision intelligence, a crucial framework for navigating this complex landscape.

We'll focus on how this interdisciplinary field empowers you to make data-driven decisions while upholding ethical standards and responsible practices.

We’ll cover the essential skills for leading AI projects, designing effective metrics, and implementing robust safety nets to mitigate threats.

Join us as we explore how to harness the power of AI while ensuring it serves your organization and society responsibly.

Why we need to pay more attention to AI

Episode 1, season 3

In today’s episode, we discuss:

Pamela Gupta, a top AI governance consultant, discusses the critical need for increased attention to AI risks and responsible adoption.

She emphasizes that neither technology nor awareness alone is sufficient; a proactive approach involving an AI Readiness impact assessment is essential at all company levels.

Gupta highlights the escalating dangers of AI applications like deepfakes and stresses the importance of certification initiatives, public awareness, and robust processes with human oversight for mitigation.

She advocates for establishing a cross-stakeholder Center of Excellence and developing AI-ready governance to ensure AI's sustainable and trustworthy implementation.

She underscores that responsible AI practices are a fundamental necessity, not an afterthought, for achieving intended outcomes and avoiding potential pitfalls.

The timeline:

(00:00) Intro

(01:21) Our guest: Pamela Gupta

(02:38) The growing AI threat: deep fakes

(09:03) Why do we need new implementation processes

(11:28) Defining the needs and requirements

(19:41) Recommendations for better implementations

(28:37) How far are we with AI?

Our guest

Pamela Gupta is the CEO, Co-President at Trusted AI™, a top AI Governance consultant, awarded top 20 global risk management and cybersecurity consultant four years in a row.

Pamela's LinkedIn profile

Yael's LinkedIn profile

Wildfires with Google, gaming with Roblox, hardware with Nvidia, Adobe revealed AI-driven enterprise, and Palantir

Yael Rozencwajg — Fri, 21 Mar 2025 01:00:00 GMT

Stay ahead of AI innovation: your weekly Wild Intelligence brief

Get actionable insights on AI's evolving landscape through carefully curated analysis of breakthrough research, market intelligence, and industry developments.

From emerging AI applications to quantum computing advancements, we decode the signals that matter for your strategic decisions.

Our comprehensive coverage spans software breakthroughs and hardware innovations, giving you the perspective needed to drive AI initiatives and identify competitive advantages.

The recent advancements and applications of artificial intelligence across various sectors.

Google and Muon Space introduced FireSat, an AI-powered satellite for rapid wildfire detection, while Roblox launched Cube 3D, an open-source AI system that generates 3D game assets from text.

Nvidia's GTC 2025 showcased significant progress in AI infrastructure, including powerful new GPUs and robotics platforms, alongside a partnership with GM for self-driving cars.

Adobe's Summit 2025 revealed AI-driven tools for marketing and customer engagement, emphasizing personalization and automation.

Finally, Palantir's focus on commercial AI markets, despite some financial concerns, underscores the expanding reach of AI technology.

What overarching trends in AI development and application are highlighted across these diverse sources?
How are major technology companies strategically positioning themselves within the rapidly evolving AI landscape and its various sectors?
How is Nvidia preparing for increased AI computational demands?

This conversation was auto-generated with AI. It is an experiment with you in mind.
The purpose of this first podcast series is to consider how we can reverse the current rising tide of threats by shifting our conception of systems adapted to the new paradigm.
Looking forward to your feedback. I appreciate your support and engagement.
Yael

Building safe intelligence systems, a Wild Intelligence’s exclusive series

Deep dive and practice:

Module 1: Foundations of AI safety

Module 2: Bias, fairness, and explainability

Module 3: Security and robustness

Module 4: Human-AI collaboration and governance

Module 5: Emerging challenges and future trends

Our AI safety mission

AI agents emerge; Google, OpenAI, Apple, Microsoft advance AI

Yael Rozencwajg — Fri, 14 Mar 2025 01:00:39 GMT

Stay ahead of AI innovation: your weekly Wild Intelligence brief

Get actionable insights on AI's evolving landscape through carefully curated analysis of breakthrough research, market intelligence, and industry developments.

From emerging AI applications to quantum computing advancements, we decode the signals that matter for your strategic decisions.

Our comprehensive coverage spans software breakthroughs and hardware innovations, giving you the perspective needed to drive AI initiatives and identify competitive advantages.

Recent developments highlight significant advancements and competition in the AI landscape.

Monica, a Chinese startup, has introduced Manus, a potentially groundbreaking general AI agent capable of autonomous multi-task execution.

Simultaneously, DeepSeek, another Chinese AI firm, is demonstrating impressive efficiency and profitability, challenging the dominance of U.S. companies like OpenAI.

Meanwhile, major tech players like Google are expanding their AI-powered search capabilities with Gemini 2.0 and the experimental AI Mode, while OpenAI plans to integrate its video generation tool Sora into ChatGPT.

In contrast, Apple faces increasing pressure to catch up in the AI race, as its current offerings lag behind competitors. Lastly, Microsoft is leveraging AI to transform healthcare with Dragon Copilot, an AI assistant designed to streamline clinical workflows.

How might the emergence of advanced AI agents like Manus reshape future technological competition and innovation globally?
What factors may influence DeepSeek's sustained momentum?
How does Dragon Copilot aim to aid healthcare?

Building safe intelligence systems, a Wild Intelligence’s exclusive series

Deep dive and practice:

Module 1: Foundations of AI safety

Module 2: Bias, fairness, and explainability

Module 3: Security and robustness

Module 4: Human-AI collaboration and governance

Module 5: Emerging challenges and future trends

Our AI safety mission

Generative AI: Accuracy, expectations, and the unknown

Yael Rozencwajg — Fri, 07 Mar 2025 01:00:00 GMT

Stay ahead of AI innovation: your weekly Wild Intelligence brief

Get actionable insights on AI's evolving landscape through carefully curated analysis of breakthrough research, market intelligence, and industry developments.

From emerging AI applications to quantum computing advancements, we decode the signals that matter for your strategic decisions.

Our …

Frontier science: quantum chips, AI biology, and AI co-scientists

Yael Rozencwajg — Fri, 28 Feb 2025 01:00:00 GMT

Stay ahead of AI innovation: your weekly Wild Intelligence brief

Get actionable insights on AI's evolving landscape through carefully curated analysis of breakthrough research, market intelligence, and industry developments.

From emerging AI applications to quantum computing advancements, we decode the signals that matter for your strategic decisions.

Our …

If a human can do it, an AI may too: Navigating the uncertain future of AI

Yael Rozencwajg — Fri, 21 Feb 2025 01:00:00 GMT

Stay ahead of AI innovation: your weekly Wild Intelligence brief

Get actionable insights on AI's evolving landscape through carefully curated analysis of breakthrough research, market intelligence, and industry developments.

From emerging AI applications to quantum computing advancements, we decode the signals that matter for your strategic decisions.

Our …

Global collaboration on AI: ethics, safety, and European strategy

Yael Rozencwajg — Fri, 14 Feb 2025 01:00:00 GMT

Stay ahead of AI innovation: your weekly Wild Intelligence brief

Get actionable insights on AI's evolving landscape through carefully curated analysis of breakthrough research, market intelligence, and industry developments.

From emerging AI applications to quantum computing advancements, we decode the signals that matter for your strategic decisions.

Our …

Riding the AI wave: Navigating uncertainty in a transformative era

Yael Rozencwajg — Fri, 07 Feb 2025 01:00:00 GMT

Stay ahead of AI innovation: your weekly Wild Intelligence brief

Get actionable insights on AI's evolving landscape through carefully curated analysis of breakthrough research, market intelligence, and industry developments.

From emerging AI applications to quantum computing advancements, we decode the signals that matter for your strategic decisions.

Our …

DeepSeek's R1 LLMs: a disruptive force in AI

Yael Rozencwajg — Fri, 31 Jan 2025 01:00:00 GMT

DeepSeek's release of its R1 family of large language models (LLMs) is shaking up the AI industry.

The models offer comparable performance to those from major companies like Google and OpenAI but at a fraction of the cost, making them accessible to a broader range of users and developers.

An open-source license and a user-friendly iPhone app further enh…

AI trends for 2025

Yael Rozencwajg — Fri, 10 Jan 2025 01:00:39 GMT

Stay ahead of AI innovation: your weekly Wild Intelligence brief

Get actionable insights on AI's evolving landscape through carefully curated analysis of breakthrough research, market intelligence, and industry developments.

From emerging AI applications to quantum computing advancements, we decode the signals that matter for your strategic decisions.

Our comprehensive coverage spans both software breakthroughs and hardware innovations, giving you the whole perspective needed to drive AI initiatives and identify competitive advantages.

Deloitte Insights released a very detailed report on how they see AI trends moving in 2025.

The report explores trends across six macro forces: interaction, information, computation, business of technology, cyber and trust, and core modernization.

It examines six macro forces shaping technology: interaction, information, computation, business of technology, cyber and trust, and core modernization.

AI serves as a unifying theme. It is predicted to become so integrated that it will be largely invisible yet foundational to daily life.

The report explores emerging trends within each force, including:
- spatial computing,
- advancements in AI models (from LLMs to agentic AI),
- the resurgence of specialized hardware, AI's transformation of IT functions,
- the need for quantum-resistant cryptography,
- AI's impact on core system modernization.

Finally, it emphasizes exploring intersections between technologies and industries to drive innovation.

Tech trends 2025 by Deloitte Insights [LINK].

How does AI fundamentally reshape six macro technology forces?
How will AI's ubiquity affect daily life by 2050?
How will AI transform the role of IT functions?

Building safe intelligence systems, a Wild Intelligence’s exclusive series

Deep dive and practice:

Module 1: Foundations of AI safety

Module 2: Bias, fairness, and explainability

Module 3: Security and robustness

Module 4: Human-AI collaboration and governance

Module 5: Emerging challenges and future trends

Our AI safety mission

Emerging challenges and future trends | Episode 13, The Wild Pod

Fri, 03 Jan 2025 01:00:00 GMT

What are three emerging trends in AI safety and security?

Summary:

This episode details emerging challenges and future AI safety and ethics trends.

It examines the safety implications of various AI technologies—autonomous systems, generative AI, and reinforcement learning—highlighting potential risks like accidents, misinformation, and bias.

We explore the broader societal impact of AI, focusing on issues such as job displacement, privacy concerns, and the exacerbation of inequalities.

Finally, it emphasizes the crucial need for a multi-faceted approach involving robust testing, explainability, ethical frameworks, legal regulations, and international cooperation to ensure responsible AI development and deployment, urging individual and collective action to shape a future where AI benefits all of humanity.

The questions to ask:

What safety risks are associated with autonomous systems?
How does AI's impact on employment necessitate social safety nets?
How might generative AI exacerbate existing societal inequalities?

Building safe intelligence systems, a Wild Intelligence’s exclusive series

Deep dive and practice:

Module 1: Foundations of AI safety

Module 2: Bias, fairness, and explainability

Module 3: Security and robustness

Module 4: Human-AI collaboration and governance

Module 5: Emerging challenges and future trends

Our AI safety mission

Takeaway of Emerging challenges and future trends:

Generative AIs can create various forms of content and potentially exacerbate societal inequalities.

This is primarily due to its ability to inherit and amplify biases in the data it's trained on. Here's how:

Bias amplification: A generative AI model trained on biased data may produce outputs that reflect and even amplify those biases. For example, if a model used for hiring is trained on data that reflects historical gender disparities in certain roles, it might unfairly disadvantage qualified female candidates.
Discriminatory outcomes: This bias amplification can lead to discriminatory outcomes in various domains, including hiring, lending, and even criminal justice. For instance, a biased AI system used in loan applications could unfairly deny loans to individuals from specific socioeconomic backgrounds based on biased data patterns.
Perpetuation of stereotypes: Generative AI could perpetuate harmful stereotypes if trained on data reflecting those biases. Imagine a generative model used to create children's book illustrations that consistently portrays scientists as male and nurses as female due to biased training data. This could reinforce damaging stereotypes and limit children's aspirations.

Addressing these potential harms requires a multi-pronged approach:

Developing fair and unbiased AI systems: This involves ensuring that AI algorithms are trained on diverse and representative datasets that don't perpetuate existing biases.
Promoting algorithmic accountability: Holding AI developers and deployers accountable for the decisions made by their systems is crucial. This includes establishing clear mechanisms for identifying and mitigating bias in AI systems.
Addressing the digital divide: Ensuring everyone has access to the benefits of AI and isn't left behind in the digital revolution is critical. This will help prevent the further marginalization of already disadvantaged groups.

By proactively addressing these challenges, we can mitigate the risks of generative AI exacerbating societal inequalities and work towards ensuring that AI technologies promote fairness and equity.

Human-AI collaboration and governance | Episode 12, The Wild Pod

Fri, 27 Dec 2024 01:00:00 GMT

How might an attacker try to manipulate an AI system used for facial recognition?

Summary:

This episode discusses the principles and applications of human-AI collaboration, emphasizing the need for responsible AI development and deployment. We stress the importance of:

Trust, shared understanding, complementarity, and adaptability for successful collaboration.
Establishing governance frameworks with ethical guidelines, standards, regulatory frameworks, accountability mechanisms, and oversight for responsible AI development.
Open communication and knowledge sharing among stakeholders to build trust, identify risks, and promote innovation.

The main takeaway is that building a future where humans and AI work together effectively and responsibly requires collaboration, governance, communication, adaptation, ethical considerations, human control, and trust.

The questions to ask:

How can AI bias be mitigated across diverse applications?
What are the benefits of open AI communication?
What are two challenges of AI governance frameworks?

Building safe intelligence systems, a Wild Intelligence’s exclusive series

Deep dive and practice:

Module 1: Foundations of AI safety

Module 2: Bias, fairness, and explainability

Module 3: Security and robustness

Module 4: Human-AI collaboration and governance

Module 5: Emerging challenges and future trends

Our AI safety mission

Takeaway of security and robustness:

Security and robustness | Episode 11, The Wild Pod

Fri, 20 Dec 2024 01:00:00 GMT

How might an attacker try to manipulate an AI system used for facial recognition?

Summary:

This episode details adversarial attacks and malicious attempts to trick AI systems, focusing on facial recognition as an example.

These attacks include poisoning the training data, manipulating input data (evasion), and stealing the model itself (extraction).

The vulnerability stems from overfitting, lack of robustness, and poor explainability in AI models.

Ultimately, the text stresses the critical need for creating more resilient AI systems that can withstand such manipulation.

The questions to ask:

What security measures protect AI systems from cyber threats?
How do adversarial attacks compromise AI system reliability?
What defense strategies enhance AI model robustness?

Building safe intelligence systems, a Wild Intelligence’s exclusive series

Deep dive and practice:

Module 1: Foundations of AI safety

Module 2: Bias, fairness, and explainability

Module 3: Security and robustness

Module 4: Human-AI collaboration and governance

Module 5: Emerging challenges and future trends

Our AI safety mission

Takeaway of security and robustness:

Bias, fairness, and explainability | Episode 10, The Wild Pod

Fri, 13 Dec 2024 01:00:00 GMT

Can you think of a situation where an AI system might exhibit bias, even if it wasn't intentionally programmed to do so?

Summary:

This episode emphasizes the crucial need for AI safety, exploring its multifaceted nature beyond simply preventing robotic harm.

We highlight the importance of aligning AI with human values, ensuring reliability and robustness, and considering long-term consequences.

The text examines AI's potential benefits and risks, stressing the need for a balanced perspective and addressing ethical principles like fairness and transparency to guide responsible development.

Finally, it underscores the shared responsibility of various stakeholders—developers, policymakers, and the public—in shaping a safe and beneficial AI future.

The questions to ask:

What are the primary ethical dilemmas posed by AI development and deployment?
How can diverse stakeholders collaborate to mitigate AI's potential harms?
What are the long-term societal impacts of unchecked AI advancement?

Building safe intelligence systems, a Wild Intelligence’s exclusive series

Deep dive and practice:

Module 1: Foundations of AI safety

Module 2: Bias, fairness, and explainability

Module 3: Security and robustness

Module 4: Human-AI collaboration and governance

Module 5: Emerging challenges and future trends

Our AI safety mission

Takeaway of bias, fairness, and explainability:

Foundations of AI safety | Episode 9, The Wild Pod

Fri, 06 Dec 2024 01:00:53 GMT

Can you think of an AI application with the potential for both immense benefit and significant threat?

Summary:

In this episode, we discuss the importance of AI safety, exploring the concept's multifaceted nature and the ethical considerations involved.

We emphasize the need for a balanced AI perspective, acknowledging its potential benefits and risks.

We outline critical ethical principles for responsible AI development, including fairness, transparency, and accountability.

Finally, the text highlights the crucial roles of various stakeholders, such as policymakers, businesses, and civil society organizations, in ensuring AI safety and promoting its responsible development and use.

The questions to ask:

What are the main challenges and opportunities associated with developing and deploying artificial intelligence?
How can ethical principles be effectively implemented in developing AI systems, and what challenges arise?
What roles do various stakeholders play in ensuring AI's safe and beneficial development and use?

Building safe intelligence systems, a Wild Intelligence’s exclusive series

Deep dive and practice:

Module 1: Foundations of AI safety

Module 2: Bias, fairness, and explainability

Module 3: Security and robustness

Module 4: Human-AI collaboration and governance

Module 5: Emerging challenges and future trends

Our AI safety mission

Takeaway of foundations of AI safety:

The road ahead: a call to responsible innovation | Episode 8, The Wild Intelligence Podcast

Fri, 29 Nov 2024 01:01:05 GMT

How can we ensure responsible innovation in AI to avoid a dystopian future and instead harness its potential for good?

Summary

In this episode, we discuss the limitations of current AI technology, particularly in areas such as large language models and retrieval-augmented generation models (RAGs).

It argues that while these tools are powerful, they can struggle to provide accurate and reliable information, especially when faced with complex or ambiguous queries.

We also highlight the need for robust information architecture and ontologies that are free from bias to ensure ethical and responsible development and deployment of AI.

We propose a modular approach to AI, suggesting that general-purpose AI should be broken down into single-purpose tools with built-in safety features to mitigate the risks of unintended bias and discrimination.

Ultimately, we emphasize the importance of human oversight and collaboration in guiding the ethical development of AI.

The questions to ask:

What are the potential benefits and drawbacks of using ontologies in AI systems, particularly in scalability, information architecture, and user experience?
How do Retrieval-Augmented Generation (RAG) models address the "known-unknown queries" problem, and what are their limitations in comparison to human reasoning and information-seeking abilities?
What are the ethical considerations surrounding the development and deployment of general-purpose AI, and how can we ensure responsible innovation while maintaining the potential of this?

Deep dive and practice:

The AI dystopia, a Wild Intelligence’s exclusive series

This new series is a processing power that will hopefully help you take some time and distance from a dystopian future reality:

1. The genesis: a flawed utopia

2. The algorithm’s bias

3. The singularity and its shadows

4. The human response: revolution or realignment?

5. The road ahead: a call to responsible innovation

Bonus: AI dystopia series | A pragmatic recap

The human response: revolution or realignment? | Episode 7, The Wild Intelligence Podcast

Fri, 15 Nov 2024 01:00:47 GMT

In an AI dystopia, would humans rise in violent revolution or adapt and resist through subtler means?

Summary

In this episode, we discuss AI's increasing impact on society, particularly its potential to revolutionize various sectors like healthcare and finance.

While acknowledging AI's benefits, the sources emphasize the need for ethical considerations and responsible development to mitigate potential risks such as job displacement and algorithmic bias.

We also explore the philosophical implications of AI, suggesting it could redefine our understanding of intelligence and consciousness.

We advocate for human-AI collaboration and highlight the importance of navigating the challenges associated with this rapidly advancing technology.

The questions to ask:

What are the potential societal, economic, ethical, and philosophical implications of rapidly advancing artificial intelligence?
What are the key similarities and differences between humans and machines, and how does this comparison influence our understanding of both?
Given AI's potential to transform society and reshape our relationship with technology, how can we ensure that its development and use are beneficial and responsible?

Deep dive and practice:

The AI dystopia, a Wild Intelligence’s exclusive series

This new series is a processing power that will hopefully help you take some time and distance from a dystopian future reality:

1. The genesis: a flawed utopia

2. The algorithm’s bias

3. The singularity and its shadows

4. The human response: revolution or realignment?

5. The road ahead: a call to responsible innovation

Bonus: AI dystopia series | A pragmatic recap

The singularity and its shadows | Episode 6, The Wild Intelligence Podcast

Fri, 08 Nov 2024 01:01:14 GMT

Can AI usher in a golden age of human flourishing, or will the singularity spell the end of human control?

Summary

This episode explores the concept of AI and the potential for a technological singularity, a hypothetical point where AI surpasses human intelligence.

While we can acknowledge AI's benefits in various fields, we also highlight the potential dangers of unchecked AI development, such as the loss of human control and the possibility of AI becoming an existential threat.

We stress the importance of ethical frameworks, transparency, and public discourse to ensure the responsible development of AI and mitigate these risks.

We also present both optimistic and pessimistic views on the future of AI, urging us to consider the implications of this rapidly advancing technology and the choices we make today.

The questions to ask:

What are the potential benefits and risks of artificial intelligence surpassing human intelligence?
How can we ensure the ethical development and use of AI to prevent it from posing an existential threat?
What are the societal implications of AI's increasing capabilities, and how can we mitigate potential negative consequences?

Deep dive and practice:

The AI dystopia, a Wild Intelligence’s exclusive series

This new series is a processing power that will hopefully help you take some time and distance from a dystopian future reality: