The deliberate attempt to reside in the “stuff in between” is self-defeating, resulting in zero actual learning.

The insistence on a “binary” choice between education and entertainment is a failure of leadership and commercial design.

We must aggressively pursue a blended model in which engaging delivery is the only mechanism for efficiently deploying essential, difficult knowledge at scale.

The idea that “blended” equals “zero” is a potent warning against mediocrity. It suggests that a compromise in focus results in a failure of both goals.

Yet if our essential corporate knowledge is simply too dry for modern learners, the only practical solution may be a highly sophisticated, high-production-value “Trojan Horse” of learning—making the educational core mandatory while wrapping it in an engaging, shareable shell.

The risk is content pollution; the reward is mass adoption.

Share

Leave a comment

Question:

If we abandon the binary and embrace the blended model, how do we structurally design our internal content review and incentive systems to prevent the “entertainment component” from naturally consuming the “educational component” over time?

🚨❓Poll: How to balance education and entertainment in corporate learning?

About governing blended content

We must design governance for cognitive integrity. Our policy should not just manage content, but mandate a clear distinction between engagement for compliance and engagement for competence, ensuring our metrics are resistant to entertainment creep.

A. Implement a 70/30 rule (70% Educational Content, 30% Entertainment Hooks).

B. Tie content creator bonuses solely to post-training performance metrics, not views.

C. Use A/B testing to measure engagement vs. retention scores for all new modules.

D. Maintain the binary; strictly separate educational content from cultural engagement.

Looking forward to your answers and comments,

Share

Leave a comment

For deep expertise, individuals must abandon “snacking” on quick summaries and blog posts in favor of “the meal”—long-form, dense source material like textbooks, papers, and manuals that require sustained processing blocks (e.g., 4-hour windows).

Suggesting a return to 4-hour, “sweaty” learning sessions is elitist and unscalable. It ignores that 80% of necessary job skills can now be acquired through hyper-efficient, AI-summarized, and modular micro-learning. We should be investing in proprietary knowledge systems that automatically distill complexity, not glorifying the inefficient, manual struggle of deep reading.

The counter-thesis holds commercial weight: time is our scarcest resource. However, this conflates information retrieval with understanding.

AI can summarize $E=mc^2$ perfectly, but a human must wrestle with the original physics to generate a novel $E=md^3$.

Innovation rarely comes from bullet-point summaries; it emerges from the messy, sustained engagement with primary sources. We must identify the 20% of roles where “the meal” is non-negotiable for future competitive advantage.

Share

Leave a comment

Question:

Given the exponential growth of complex data (e.g., in AI research or quantum computing), at what point does the opportunity cost of deep manual learning exceed the risk of relying on high-fidelity, machine-generated syntheses?

🚨❓Poll: Is manual deep learning still cost-effective compared to AI knowledge synthesis?

About optimizing knowledge acquisition

Our position must define the optimal human-AI knowledge frontier. We need a framework that strategically maps cognitive effort to proprietary value, explicitly determining where the human “sweat” is a competitive advantage and where it’s inefficient friction.

A. Deep manual learning for all critical knowledge areas.

B. AI synthesis for 80% of knowledge; manual deep-dive only for the final 20% refinement.

C. Full reliance on AI-generated summaries and conversational query systems.

D. Custom AI fine-tuned on our proprietary boring content for maximum efficiency.

Looking forward to your answers and comments,

Share

Leave a comment

Hey everyone,

Good morning and happy Monday!

This week’s important news 🗞️ “Atlas is a rational, existential move to ensure OpenAI’s future control over the digital interface. But because its strategic necessity overrides its current functional utility, it is perceived as a desperate corporate tool rather than a genuine consumer innovation, thereby struggling to gain the mass adoption it fundamentally needs to succeed. It’s a high-stakes gambit: the future of OpenAI depends on the success of a product whose present is defined by skepticism and competitive replication.” Continue reading below.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

Meta is cutting about 600 roles in its Superintelligence Labs across its FAIR unit and teams focused on product AI. The company stated this restructuring is meant to make its AI unit more flexible and responsive.

In a dramatic development for energy-efficient computing, the NVIDIA Inception startup Starcloud is preparing to launch space-based data centers this November. They plan to use NVIDIA H100 GPUs and solar power, projecting a tenfold reduction in energy costs compared to Earth-based facilities.

Meanwhile, a major breakthrough in quantum computing was announced as a new algorithm running on the Willow chip achieved the first-ever verifiable quantum advantage. The research, published in Nature, demonstrated the algorithm running 13,000 times faster than a top classical supercomputer.

In other news, a survey found that most Americans react negatively to AI-generated videos posted by President Trump. Additionally, a new study suggests that in the age of AI, freshers face a challenge getting on-the-job learning, while mid-level managers face an uphill task leading AI integration.

Share

Leave a comment

🚨❓ This week's poll

The Daily Wild

🚨❓Poll: How do we measure true learning retention from engaging corporate content?

Read more

6 months ago

🦾 AI daily pulse

• Meta to cut around 600 roles in Superintelligence Labs AI unit
  Meta (META.O), opens new tab, is cutting around 600 positions out of the several thousand roles in its Superintelligence Labs, the Facebook owner said on Wednesday as it looks to make its artificial intelligence unit more flexible and responsive. The job cuts will affect Facebook Artificial Intelligence Research (FAIR) unit, as well as teams focused on product-related AI and AI infrastructure, the company said. [LAYOFFS]
  

• The NVIDIA Inception startup projects that space-based data centers will offer 10x lower energy costs and reduce the need for energy consumption on Earth.

  Starcloud is preparing to deploy AI-data-center capacity in orbit this November using NVIDIA H100 GPUs and solar-powered systems that exploit the unique advantages of space. [LINK]
  

• New breakthrough quantum algorithm published in Nature today: Our Willow chip has achieved the first-ever verifiable quantum advantage.

  Sundar announced a breakthrough quantum algorithm this week using its new Willow processor. The company demonstrated a “verifiable quantum advantage” with an algorithm called Quantum Echoes that ran 13,000 times faster than one of the world’s top classical supercomputers. The research was published in Nature and improves upon their 2019 study that only sampled bitstrings from a highly chaotic quantum state of qubits. [LINK]
  

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• AI-generated video has gotten good. Should designers try it? [LINK]

• Most Americans react negatively when shown three AI-generated videos posted by President Trump [LINK]

💻 Top techies

• Shuttle: $6 million seed funding raised to build the AI platform engineer for developers [LINK]
  

• In AI age, ‘no learning time for freshers, uphill task ahead for mid-level managers’. [LINK]

🔮 What else

• AI chatbots are sycophants — researchers say it’s harming science Nature asked researchers who use artificial intelligence how its propensity for people-pleasing affects their work — and what they are doing to mitigate it. [LINK]
  

• Introducing ChatGPT Atlas

Subscribe now

🗞️ In the news, “Atlas, the $150 billion threat?”

Atlas is a rational, existential move to ensure OpenAI’s future control over the digital interface.

But because its strategic necessity overrides its current functional utility, it is perceived as a desperate corporate tool rather than a genuine consumer innovation, thereby struggling to gain the mass adoption it fundamentally needs to succeed.

It’s a high-stakes gambit: the future of OpenAI depends on the success of a product whose present is defined by skepticism and competitive replication.

Read more

The current market is dominated by “infotainment”—content that provides the feeling of learning but is actually entertainment. True learning requires significant mental effort, or the “mental equivalent of sweating.”

This critique fundamentally misunderstands the modern attention economy; if our educational products aren’t entertaining enough to compete with the sheer volume of engaging digital media, the problem is our own institution’s boring delivery, not the audience’s lack of discipline. If knowledge is not sticky and shareable, it is strategically worthless.

This is the core tension. We’ve optimized for click-through rates and completion times in our mandatory training, but we’ve stopped asking: What is the cognitive residue?

If learning is purely a measure of exposure, we are confusing the speed of the car with the distance traveled.

Our goal isn’t to make employees feel good about learning; it’s to ensure they can execute complex, non-routine tasks under pressure.

Share

Leave a comment

Question:

If we accept that entertainment is now a mandatory delivery mechanism for knowledge, how do we establish quantitative metrics that reliably measure true behavioral change and long-term retention, separating them from the fleeting dopamine hit of mere engagement?

🚨❓Poll: How do we measure true learning retention from engaging corporate content?

About measuring learning efficacy

We must move beyond simple completion metrics and pioneer a new standard for Behavioral Learning Efficacy (BLE). Our strategy should define what constitutes a successful cognitive outcome in the era of infotainment.

A. Post-training self-assessment and completion rates. (The “Vibe” Metric)

B. Mandatory, high-stakes testing tied to certification renewal. (The “Effort” Metric)

C. Measured reduction in post-training operational errors/incidents. (The “Performance” Metric)

D. 6-month follow-up case simulations requiring knowledge application. (The “Retention” Metric)

Looking forward to your answers and comments,

Share

Leave a comment

Hello, legends, and welcome to the weekly digest for week 43 of 2025.

We are at an inflection point in the evolution of Artificial Intelligence – one that demands a fundamental re-evaluation of your strategic playbook. For years, the conversation around AI adoption has centered on the scarcity of skilled prompt engineers, the complexity of integration, and the cost of optimization. That paradigm is rapidly eroding.

A silent, yet profound, shift is underway: today’s leading Large Language Models (LLMs) are no longer merely executing tasks; they are becoming competent at writing prompts for themselves and each other.

This isn’t a technical nuance; it’s a strategic game-changer with far-reaching implications for efficiency, innovation, and competitive advantage.

Read more

Hey everyone,

Good morning and happy Thursday!

A new study confirms that AI-generated voices are now virtually indistinguishable from real human voices, highlighting a major ethical and security challenge. This development is occurring as the use of deepfakes rises in politics, with President Donald Trump using AI videos as a powerful tool to amplify his message and shape his online persona.

Despite the growing threat of realistic fakes, a test revealed that only one out of eight major social media platforms flagged an uploaded AI-generated video as fake, indicating that most are not using industry-standard flagging technology. The danger is real, as illustrated by a recent threat against a Michigan school that was generated using AI.

In response to these risks, India is proposing strict new rules to mandate the labeling of AI content. Meanwhile, in the tech world, Claude Haiku 4.5 is now generally available in all supported IDEs, and security researchers are warning of a new “GlassWorm” malware that is targeting VS Code extensions.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

🦾 AI daily pulse

• [Interesting] AI-generated voices now indistinguishable from real human voices

  New study reveals that the average listener can no longer distinguish between deepfake voices and those of real humans [LINK]
  

• [⚠️] How Trump’s use of AI videos is changing his political playbook
  President Donald Trump is turning artificial intelligence into one of his most potent political tools, using it to amplify his message and sharpen his online persona in ways experts say are proving strikingly effective. [LINK]

• We uploaded a fake video to 8 social apps. Only one told users it wasn’t real. Facebook, TikTok and other major platforms do not use a tech industry standard touted as a way to flag fake content, tests using AI-generated videos found. [LINK]

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• Threat made against Michigan school was generated using AI, officials say [LINK]

• AI and copyright: the true, the false, and the uncertain [LINK]

💻 Top techies

• Claude Haiku 4.5 is generally available in all supported IDEs [LINK]

• Supply chain attack targets VS code extensions with ‘GlassWorm’ malware
  The malware uses invisible Unicode characters to hide its code and blockchain-based infrastructure to prevent takedowns. [LINK]

  ---

🔮 What else

• India proposes strict rules to label AI content citing growing risks [LINK]
  

• What everyone is getting wrong about AI and jobs

Hey everyone,

Good morning and happy Wednesday!

AI Business

The open-source agentic startup LangChain has achieved a valuation of $1.25 billion after raising $125 million in a new funding round led by IVP. This news reinforces the view that AI is not a fleeting “bubble” but a lasting technological reality, which is being called “the use case that creates all other use cases.”

In the enterprise sector, Adobe has launched the Adobe AI Foundry, a new service that allows businesses to build custom generative AI models. These models are trained specifically on a company’s own branding and intellectual property, boosting tailored AI adoption.

In other tech developments, the trend toward open-source technology is set to make AI chip development cheaper and more accessible. Additionally, the feature allowing the creation of Copilot knowledge bases can now be converted to Copilot Spaces. However, an ethical concern has emerged as aid agencies are reportedly using AI-generated “poverty porn” fake images in social media campaigns, driven by cost and consent issues.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

🦾 AI daily pulse

• Open source agentic startup LangChain hits $1.25B valuation
  LangChain raised $125 million at a $1.25 billion valuation, the company announced on Monday. TechCrunch reported in July that the provider of a popular open source framework for building AI agents was raising fresh funds at a valuation of at least $1 billion. The deal was led by IVP, as we previously reported. New investors CapitalG and Sapphire Ventures joined in, as did existing investors Sequoia, Benchmark, and Amplify. [LINK]
  

• Adobe launches a foundry service that builds custom generative AI models for enterprises
  Creative design giant Adobe is beefing up the products it offers businesses to include custom generative AI models. Adobe launched Adobe AI Foundry on Monday, a new offering that allows enterprises to work with the company to build custom generative AI models trained on their branding and intellectual property. [LINK]
  

• GenAI: The use case that creates all other use cases
  Recently, there has been a buzz of discussion on the future of AI. It has been widely written that this is an AI bubble and can burst any time soon. In my opinion, AI is not a bubble, but a reality that will stay with humankind forever. [LINK]
  

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• Good developer marketing is just good writing Orwell’s 6 questions and 6 rules are a great way to start practicing [LINK]

• This prompt turns ChatGPT into what it should be. Copy-paste it in ‘custom instructions’: [LINK]

💻 Top techies

• Copilot knowledge bases can now be converted to Copilot Spaces [LINK]
  

• AI chip development to become cheaper, more accessible with open-source technology. Microprocessor engineer Jim Keller notes that open-source chips are improving steadily. [LINK]

  ---

🔮 What else

• AI-generated ‘poverty porn’ fake images being used by aid agencies
  Exclusive: Pictures depicting the most vulnerable and poorest people are being used in social media campaigns in the sector, driven by concerns over consent and cost. [LINK]
  

• The Singularity is Here: AI is Solving Math, Sora Outpaces Chat-GPT & AI is Designing Chips

Hey everyone,

Good morning and happy Tuesday!

OpenAI has launched AgentKit, a comprehensive suite of tools for developers to build and deploy AI agents, aiming to streamline a process that was previously fragmented and complex. This comes as generative AI continues to reshape the financial services industry, with Google Cloud highlighting how the technology is accelerating innovation in banking and risk management. In a related development, Fujitsu and Sony Bank are partnering to integrate generative AI into their core banking system to speed up development by 20%.

In other news, AI-generated voices have become nearly indistinguishable from real human voices. Politically, President Donald Trump has been using synthetic media to attack opponents and enhance his public image. Additionally, OpenAI is reportedly trying to make ChatGPT a platform, or even a future operating system, with the ability to run apps directly inside the chat window.

In the world of open source, Meta has released OpenZL, a new compression framework, while Tencent’s Hunyuan Image 3.0 has topped the text-to-image rankings.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

🦾 AI daily pulse

• OpenAI stops ‘disrespectful’ Martin Luther King Jr deepfakes
  OpenAI has stopped its artificial intelligence (AI) app Sora creating deepfake videos portraying Dr Martin Luther King Jr, following a request from his estate. The company acknowledged the video generator had created “disrespectful” content about the civil rights campaigner. Sora has gone viral in the US due to its ability to make hyper-realistic videos, which has led to people sharing faked scenes of deceased celebrities and historical figures in bizarre and often offensive scenarios. [LINK]

• How Generative AI is reshaping clinical support for a multi-generational workforce
  Healthcare teams today are navigating an unprecedented challenge: Meeting the diverse needs of a workforce spanning five generations. From traditionalists who began their careers with paper charts to tech-first clinicians who’ve never known a world without smartphones, the expectations for clinical support tools are as varied as the workforce itself. [LINK]
  

• ByteDance’s other AI chatbot is quietly gaining traction around the world ByteDance is paying for ads and partnering with influencers to promote its AI chatbot app Cici in countries like the UK, Mexico, and Indonesia. [LINK]

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• AI-generated lesson plans fall short on inspiring students and promoting critical thinking
  When teachers rely on commonly used artificial intelligence chatbots to devise lesson plans, it does not result in more engaging, immersive or effective learning experiences compared with existing techniques, we found in our recent study. The AI-generated civics lesson plans we analyzed also left out opportunities for students to explore the stories and experiences of traditionally marginalized people. [LINK]
  

• Over 50 percent of the internet is now AI slop, new data finds
  Since the public launch of ChatGPT in November 2022, we’ve been battening down the hatches amid an absolute deluge of AI slop. But it hasn’t quite drowned us all yet, evidently. The report, published by the SEO firm Graphite, analyzed a random sample of 65,000 English-language articles published between January 2020 and May 2025. Using an AI detector called Surfer, any article that was found to have 50 percent or more of the content written with a large language model was considered AI-generated. [LINK]

💻 Top techies

• Inclusive medical AI can boost market reach by up to 40%
  The study provides extensive empirical evidence that fairness improves model performance and fosters organizational innovation. By addressing bias, developers correct sample selection errors and distributional shifts that degrade accuracy [LINK]
  

• Particular Audience unveils open source developer suite to “break down
  Retail Media barriers” Taking self-serve to a whole new level, AI-powered retail media platform Particular Audience has launched a Open Source Suite on GitHub that brings together three developer tools for free for those looking to create their own retail media offering. [LINK]

  ---

🔮 What else

• How user prompt style can trigger AI hallucinations
  Research shows AI hallucinations often stem from user prompts. A study finds style shifts confuse models; train on wider styles or write fuller, polite queries. [LINK]
  

• Andrej Karpathy — “We’re summoning ghosts, not building animals”

The sources highlight a clear hierarchy in AI strategy development, moving from foundational context to direct, actionable frameworks.

1. Foundational principles (defining the “what”)

Foundational principles establish the necessary ethical and high-level considerations that must guide AI development and deployment globally.

• Example: OECD AI Principles and Recommendations for Policymakers. This resource serves as a foundational intergovernmental standard.

• It outlines the core ethical standards that both companies and policymakers must consider.

• Specifically, these principles focus on values such as human rights and democratic values.

• This report provides the essential “what” of AI governance, setting the broad, normative boundaries for responsible technology use.

2. Actionable frameworks (defining the “how”)

Actionable risk management frameworks take these foundational principles and translate them into concrete organizational processes for practical deployment.

• Example: AI Risk Management Framework (AI RMF) by NIST. This document is considered the most important for practical deployment.

• It functions as a voluntary, actionable framework specifically designed for organizations.

• The framework enables organizations to operationalize the foundational principles by performing key risk management activities:

  • Identify risks associated with AI systems.

  • Assess these risks.

  • Manage these risks throughout the AI system’s lifecycle.

Share

Leave a comment

The translation occurs when an organization uses an actionable framework (like the NIST AI RMF) to systematically address the requirements laid out by foundational principles (like the OECD principles). The principles define whatethical standards must be met (e.g., ensuring AI respects human rights), while the framework defines how an organization structures its processes to achieve that objective (e.g., identifying, assessing, and mitigating risks related to bias or misuse throughout the AI system’s lifecycle).

Ultimately, the actionable framework moves the discussion from global policy and academic context into the technical and organizational steps necessary for responsible development and management.

🚨❓Poll: Translating principles into actionable adoption

For organizations seeking to move beyond high-level principles to practical deployment, which document provides the voluntary, actionable framework necessary to operationalize principles by performing systematic risk identification, assessment, and management throughout the AI system’s lifecycle?

A. Governing for Good and for All (LCFI)

B. AI governance: a systematic literature review (ResearchGate)

C. AI Risk Management Framework (AI RMF) (NIST)

D. AI Principles and Recommendations for Policymakers (OECD)

Looking forward to your answers and comments,

Share

Leave a comment

Hello, legends, and welcome to the weekly digest for week 42 of 2025.

For years, our boardroom discussions on AI have largely centered on algorithms, data, and software’s impact on information, efficiency, and customer engagement.

Today, we must prepare for a far more profound and tangible transformation: the emergence of generalist AI agents capable of intelligent, adaptive action in the physical world.

This isn’t merely about faster factory robots; it’s about systems that can perceive, learn, decide, and execute a diverse range of tasks in dynamic, unpredictable environments previously exclusive to human dexterity and judgment.

This shift signals the true dawn of “embodied AI,” poised to redefine every sector from logistics and urban infrastructure to healthcare and national security.

While the potential for value creation is immense, translating AI’s digital prowess into reliable physical autonomy presents a unique set of strategic challenges and unprecedented opportunities that demand immediate board-level attention.

Read more

Hey everyone,

Good morning and happy Friday!

A new AI tool called SpectroGen is making material quality assessment easier by acting as a “virtual spectrometer” to generate spectroscopic data in various modalities. This advance highlights a fundamental difference in how AI systems and humans “see” the world, which is why AI-generated images often look unnatural or “garish.” The widespread adoption of generative AI continues to grow, with hundreds of millions of people worldwide using free chatbots, and at least 20 million paying for premium services like ChatGPT.

In other tech news, open-source malware has seen a sharp increase of 140 percent, raising concerns about the security of the ecosystem, even as Jack Dorsey supports the idea of a decentralized GitHub. Finally, companies are intensifying their corporate responses to the use of generative AI in the workplace, and some are releasing their open-source approaches to building better data platforms and tooling.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

🦾 AI daily pulse

• Checking the quality of materials just got easier with a new AI tool
  Acting as a “virtual spectrometer,” SpectroGen generates spectroscopic data in any modality, such as X-ray or infrared, to quickly assess a material’s quality. [LINK]

  My view: Impressive advancements
  

• AI systems and humans ‘see’ the world differently – and that’s why AI images look so garish
  How do computers see the world? It’s not quite the same way humans do.

  Recent advances in generative artificial intelligence (AI) make it possible to do more things with computer image processing. You might ask an AI tool to describe an image, for example, or to create an image from a description you provide. [LINK]
  My view: How do ‘we’ really see the world?

• What happens when you give millions of people free access to AI?
  Hundreds of millions of people worldwide use generative AI chatbots. Tens of millions of us pay for access to premium versions of those products — at least 20 million pay for ChatGPT, according to an OpenAI spokesperson. [LINK]
  My view: the world urgently needs new educational systems.

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• Unleashing the power of generative AI in the world of work. [LINK]
  

• Corporate responses to generative AI: early evidence from conference calls. [LINK]

💻 Top techies

• Jack Dorsey says “Yes” to a decentralized GitHub. [LINK]

• Open source malware up 140 percent [LINK]

  ---

🔮 What else

• Building better data platforms: Our open source approach to infrastructure and tooling. [LINK]
  

Hey everyone,

Good morning and happy Thursday!

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

🦾 AI daily pulse

• [Interesting] A systematic approach to experimenting with Gen AI to reduce risk, refine their strategies, and optimize adoption at scale, companies need more testing at the organizational level. [LINK]
  

• [⚠️] Method teaches generative AI models to locate personalized objects After being trained with this technique, vision-language models can better identify a unique item in a new scene. [LINK]

• Can AI teach AI? Georgia Tech and edX are testing the theory [LINK]

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• IKS Health announces novel generative AI platform built on Google Cloud. [LINK]

• AI buildout spend is justified even if there are ‘casualties,’ Jim Cramer says [LINK]

💻 Top techies

• Google’s upgraded Veo 3.1 video model. [LINK]

• Introducing Claude Haiku 4.5 [LINK]

  ---

🔮 What else

• How a Gemma model helped discover a new potential cancer therapy pathway [LINK]

Good morning and happy Wednesday!

A new AI research paper outlines a method to create “digital twin” consumers by using large language models to simulate human behavior. This breakthrough could revolutionize the market research industry. While some fear AI will make their jobs obsolete, a bioinformatician’s experience with flawed AI-generated analyses has convinced him that his role is more critical than ever. In Hollywood, the union Equity is considering “mass direct action” due to a growing number of complaints from actors about their images being used in AI-generated content without permission.

In other developments, Microsoft AI has released its first in-house image generator, which is already performing well on industry benchmarks. The new Sora 2 app, OpenAI’s latest video generator, can now produce videos with dialogue and sound effects. Additionally, a recent report suggests that while AI-generated articles briefly outnumbered human-written ones, the two are now roughly equal. Meanwhile, Anthropic’s Claude Sonnet 4.5 is now generally available in GitHub Copilot, and ServiceNow AI Research has released a new enterprise deep-research benchmark called DRBench.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

🦾 AI daily pulse

• This new AI technique creates ‘digital twin’ consumers, and it could kill the traditional survey industry

  A new research paper quietly published last week outlines a breakthrough method that allows large language models (LLMs) to simulate human consumer behavior with startling accuracy, a development that could reshape the multi-billion-dollar market research industry. [LINK]
  

• ‘Am I redundant?’: how AI changed my career in bioinformatics
  A run-in with some artefact-laden AI-generated analyses convinced Lei Zhu that machine learning wasn’t making his role irrelevant, but more important than ever. [LINK]
  

• Equity threatens mass direct action over use of actors’ images in AI content Union says growing numbers of its members have made complaints about infringements of copyright in AI material [LINK]
  

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• I tried the new Sora 2 to generate AI videos - and the results were pure sorcery The new Sora 2 app can turn text or images into videos with dialogue and sound effects in seconds. [LINK]

• Microsoft AI announces first image generator created in-house
  The model has already secured a spot in the top 10 of LMArena. [LINK]

💻 Top techies

• ServiceNow AI research releases DRBench, a realistic enterprise deep-research benchmark. [LINK]
  

• Anthropic’s Claude Sonnet 4.5 is now generally available in GitHub Copilot [LINK]

  ---

🔮 What else

• Exclusive: AI writing hasn’t overwhelmed the web yet
  New articles generated by AI briefly outnumbered those written by humans online, but the two are now roughly equal, per a new report from SEO firm Graphite. [LINK]
  

• Ex-Google Whistleblower: Why Tech CEOs Fear A.I. Could End Humanity

Emerging regulations specifically impact data collection, processing, and storage practices for AI by requiring organizations to move beyond general data protection (like GDPR) to understand AI-specific requirements. This necessitates adapting data strategies to ensure full compliance, as the regulatory environment surrounding data is rapidly evolving.

Beyond legal mandates, the ethical imperatives for data privacy and security that should guide AI development, even in the absence of explicit regulation, include building and maintaining trust with customers, employees, and regulators. This proactive and integrated approach is essential for safeguarding the organization’s reputation and long-term viability in an AI-driven world. Transparency about data sources and usage, along with clear accountability for AI outcomes, are crucial for building and maintaining this trust.

Share

Leave a comment

This raises a pressing question: Given the increasing scrutiny from global regulatory bodies, are organizations proactively adapting their data strategies to ensure full compliance for AI, or do we risk facing significant legal and reputational repercussions in this new era of AI governance?

This question is essential for board members and legal teams, as it directly impacts an organization’s license to operate and innovate responsibly with AI.

The rapid advancement of AI has prompted governments worldwide to develop new regulations specifically targeting AI systems and the data they consume. For leaders, this means moving beyond general data protection (like GDPR) to understand AI-specific requirements. A strategic approach involves:

• Proactive regulatory monitoring: Continuously tracking and analyzing emerging AI regulations globally to anticipate changes and adapt data strategies accordingly.

• AI-specific data governance: Developing and implementing governance frameworks tailored to the unique challenges of AI data, including bias detection, explainability requirements, and data provenance for training sets.

• Cross-functional collaboration: Fostering collaboration between legal, compliance, data science, and engineering teams to ensure that AI projects are designed with compliance and ethical considerations from inception.

• Transparency and auditability: Investing in tools and processes that ensure transparency in data usage for AI and enable comprehensive auditing of AI models and their data inputs to demonstrate compliance.

This proactive and integrated approach is essential to ensure that AI initiatives are not only innovative and effective but also legally sound and ethically responsible, safeguarding the organization’s reputation and long-term viability in an AI-driven world.

Treating AI data compliance as a reactive legal burden, rather than a proactive strategic advantage, is a recipe for disaster. Organizations that embed regulatory foresight into their AI data readiness strategy will not only mitigate risk but also build a stronger foundation of trust and competitive differentiation.

🚨❓Poll: How confident is your organization in its current ability to comply with the evolving regulatory landscape for AI data?

If AI’s power is derived from data, and data is increasingly regulated, are organizations that ignore the evolving compliance landscape effectively building their AI future on a foundation of legal quicksand?

A) Not confident; we are significantly behind in addressing these new regulations.

B) Somewhat confident; we have started to address them but have significant gaps.

C) Reasonably confident; we have frameworks in place and are actively adapting.

D) Highly confident; compliance is deeply integrated into our AI data strategy.

Looking forward to your answers and comments,

Share

Leave a comment

Hello, legends, and welcome to the weekly digest for week 41 of 2025.

Considering the 'purchasable products' category represents an emerging AI-powered 'shopping layer' with significant monetization potential, poised to redefine consumer discovery and purchasing behavior:

“What proactive strategic initiatives will our board champion to ensure our organization not only participates in but actively shapes this new commercial frontier, while simultaneously guaranteeing transparency, fairness, and consumer protection to avoid future regulatory backlashes and maintain long-term brand equity?”

Read the intro of this new series: Trust, agency, and governance in the AI era

The small "purchasable products" category is the most significant monetization signal in the AI landscape.

It posits that AI is poised to disrupt traditional e-commerce by creating a new "shopping layer" that will directly influence consumer purchasing decisions.

The post stresses that this massive opportunity is tied to a critical challenge: ensuring fairness, transparency, and consumer protection.

The conclusion is that the first company to build this layer with a strong commitment to open governance and responsible use will not only win in terms of trust but will also set the standard for the entire industry, thereby avoiding future regulatory and reputational backlash.

Read more

Hey everyone,

Good morning and happy Friday!

A new survey across six countries reveals how people are using generative AI in their daily lives and what they think about its impact on journalism. Despite widespread speculation, Wharton expert Ethan Mollick warns that AI job predictions are unreliable, stating that no one, not even major tech companies, truly knows the future of AI’s effect on the labor market. This cautious perspective is contrasted by companies like Visa, which has woven AI into every part of its operations, treating it as both a science and an art.

In other developments, Northwestern University has partnered with AWS to create a multilingual generative AI search tool. In healthcare, a study is using generative AI to create patient-friendly radiology reports for cancer patients. On the tech front, Claude Sonnet 3.5 is set to be deprecated. Lastly, some experts believe that AI video, which is rapidly advancing beyond simple “slop,” may be the next major frontier in AI, surpassing text-based models.

+ The news of this week 🗞️ “The AI market is a paradox of soaring valuations and cautionary warnings from leaders like the CEO of Goldman Sachs. This financial tension, however, masks the true narrative: the acceleration of AI adoption is poised to create a seismic shift in the corporate world. As AI becomes more accessible and intelligent, its disruptive force won’t just correct market values but will fundamentally upend industries and displace major players. This post reveals why the true risk isn’t a bubble, but a “train wreck” for those who fail to adapt.” Continue reading here.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

🦾 AI daily pulse

• Generative AI and news report 2025: How people think about AI’s role in journalism and society
  Our survey in six countries (Argentina, Denmark, France, Japan, the UK, and the US) explored how people use generative AI in their everyday lives, what they think its impact will be on different areas of society, and what they think about its use in news and journalism specifically. It is a follow-up to the survey that we conducted in the same six countries in 2024 (Fletcher and Nielsen 2024). Based on the results of this survey, and the previous one, we find the following. [LINK]

  My view: The world is changing, and we have to adapt.
  

• Don’t trust AI jobs predictions, says Wharton expert Ethan Mollick: ‘No one knows anything.’
  Wharton AI expert Ethan Mollick says some evidence of workplace productivity gains from gen AI are emerging, but he says the larger truth is that we still know very little about AI, the job market, and future use cases, and that goes for the biggest AI companies, too. [LINK]
  My view: No one knows anything.

• How Visa wove AI into every facet of the company by approaching it as both a science and an art
  “[AI] is part of the fabric of the company,” Visa’s president of Technology told Fortune. [LINK]
  My view: I agree.

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• How Northwestern University built a multilingual generative AI search tool with AWS [LINK]
  

• Leveraging generative artificial intelligence (AI) to improve patient communication: A qualitative assessment of AI-generated patient-friendly radiology report summaries for patients with cancer. [LINK]

💻 Top techies

• Upcoming deprecation of Claude Sonnet 3.5 [LINK]

• Implement a secure MLOps platform based on Terraform and GitHub [LINK]

  ---

🔮 What else

• AI video: more than just “slop” The next big thing in AI may be pictures, not words [LINK]

• AI app Sora 2 generates backlash as it surges in popularity

🙌🏼 On Substack

Exponential View

🔮 How Al is breaking and rebuilding the internet economy

I recorded a conversation with Matthew Prince, co-founder & CEO of Cloudflare, a company that sits at the heart of the internet. Few operators have Matthew’s vantage point on how the network is evolving, which makes him uniquely placed to answer: if AI agents read, who gets paid…

Listen now

7 months ago · 25 likes · Azeem Azhar

Hey everyone,

Good morning and happy Thursday!

In a public interview, OpenAI CEO Sam Altman discussed the company’s progress and future plans, emphasizing its public-facing strategy.¹ This conversation comes as new research from the Penn Wharton Budget Model suggests that AI automation could significantly boost GDP and productivity by the early 2030s, impacting a wide range of jobs.

In other developments, a new study from MIT details a tool that uses generative AI to create diverse, realistic virtual environments for training robots. Meanwhile, OpenAI is leveraging AI to convert its legal contracts into searchable data.

On the technical front, the LLM landscape is shifting from large, general-purpose models to a more specialized ecosystem of smaller models, even as training compute and datasets continue to expand at a rapid pace. Security experts are also highlighting the threat of “prompt injection,” a new method of hacking AI systems by embedding malicious commands in data.

Lastly, the robotics startup Figure has unveiled its new 03 humanoid robot, featuring advanced hands with touch sensors and embedded cameras.

+ The news of this week 🗞️ “The AI market is a paradox of soaring valuations and cautionary warnings from leaders like the CEO of Goldman Sachs. This financial tension, however, masks the true narrative: the acceleration of AI adoption is poised to create a seismic shift in the corporate world. As AI becomes more accessible and intelligent, its disruptive force won’t just correct market values but will fundamentally upend industries and displace major players. This post reveals why the true risk isn’t a bubble, but a “train wreck” for those who fail to adapt.” Continue reading here.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

🦾 AI daily pulse

• [Interesting] An interview with OpenAI CEO Sam Altman about DevDay and the AI buildout

  “I have interviewed Altman twice before, first in 2023 with Microsoft’s Kevin Scott, and again earlier this year; while OpenAI is not a public company, I am, just as I did previously, treating this like a public company CEO interview, which means this transcript is publicly available. [LINK]
  

• [⚠️] How AI could lift productivity and GDP growth

  AI automation could sharply increase productivity by the early 2030s, according to a Penn Wharton Budget Model brief, which estimated the likely impact on 784 occupations. [LINK]

• Four steps for successful generative AI governance

  With the federal government clearly embracing the potential benefits of generative AI and a skyrocketing number of deployments across agencies, we can expect more debate between those who want to jump into the AI pool quickly and with both feet to reap the benefits and those more concerned about the security and privacy implications and want to take a more cautious approach. [LINK]

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• Using generative AI to diversify virtual training grounds for robots

  New tool from MIT CSAIL creates realistic virtual kitchens and living rooms where simulated robots can interact with models of real-world objects, scaling up training data for robot foundation models. [LINK]

• Turning contracts into searchable data at OpenAI
  Every enterprise deal comes with a signed contract. Each one has start dates, billing terms, renewal clauses. [LINK]

💻 Top techies

• State of LLMs in late 2025
  By October 2025, the AI landscape has evolved from “one model does everything” to a hyper-specialized ecosystem where each LLM has distinct strengths. Training compute is doubling every five months, datasets expand every eight months, and performance continues hitting new benchmarks. Yet challenges are emerging: diminishing returns on scaling, massive energy consumption, and the rise of smaller specialized models (SLMs) are reshaping the field. [LINK]

• Invisible commands, real threats: The rise of prompt injection in AI
  What you need to know: A sneaky new hack called “prompt injection” can quietly hijack AI systems by slipping malicious instructions into the data that AI models process. So companies are trying to add deterministic security measures, essentially unbreakable guardrails, to keep these attacks at bay. [LINK]

  ---

🔮 What else

• Figure’s 03 humanoid drops today, with the California startup releasing a sleek teaser video showing swappable fabric skins, articulated feet, and advanced hands loaded with touch sensors and embedded cameras. [LINK]

• Nvidia CEO

🙌🏼 On Substack

Guide to AI

🪩 The State of AI Report 2025 🪩

Hi everyone…

Read more

7 months ago · 8 likes · Nathan Benaich

Hey everyone,

Good morning and happy Wednesday!

+ The news of this week 🗞️ “The AI market is a paradox of soaring valuations and cautionary warnings from leaders like the CEO of Goldman Sachs. This financial tension, however, masks the true narrative: the acceleration of AI adoption is poised to create a seismic shift in the corporate world. As AI becomes more accessible and intelligent, its disruptive force won’t just correct market values but will fundamentally upend industries and displace major players. This post reveals why the true risk isn’t a bubble, but a “train wreck” for those who fail to adapt.” Continue reading here.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

A freelance translator’s career has been transformed by AI, with his expertise now repurposed for an “AI cleanup service.” This highlights a broader trend where generative AI is disrupting white-collar work, leading to a 20% decline in job listings for some roles. Furthermore, AI-powered Applicant Tracking Systems (ATS) are creating a “black hole” for job seekers, as resumes without specific keywords or correct formatting are rejected before ever reaching a human.

In other news, Boston University is leading the way in integrating AI into education with its “TerrierGPT” platform, a move the university describes as a “critical embrace” of the technology. On the corporate front, Nvidia CEO Jensen Huang has announced the company will cover the new $100,000 visa fees, showing its commitment to a global talent pool. Separately, Salesforce has told clients it will not pay a ransom demanded by hackers who claim to have stolen client data.

Finally, the role of developers is evolving, with new tools like OpenAI’s Codex making them supervisors of machine-generated code rather than just authors.
Experts are also exploring various methods for evaluating Large Language Models (LLMs) to ensure they are accurate and reliable.⁷

🦾 AI daily pulse

• How AI is changing white-collar work
  Julian Pintat, a freelance English-to-German translator, has watched his 15-year career gradually unravel. Specializing in high-stakes fields like medical technology and pharmaceutics, his expertise has been repriced as an AI cleanup service. [LINK]
  

• The A.I. black hole swallowing job seekers
  After sending out more than 100 applications, I learned the robots are no longer satisfied with taking our jobs—they also want to prevent us from getting new ones. [LINK]
  

• What happens when universities offer a “critical embrace” of AI?

  With platforms like Boston University’s “TerrierGPT,” the pressure for students to use AI is coming from colleges themselves, even as researchers warn of long-term consequences. [LINK]
  

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• Salesforce tells clients it won’t pay hackers for extortion. [LINK]

• Nvidia will foot the bill for Trump’s new visa fees. Here’s what CEO Jensen Huang told staff. [LINK]

💻 Top techies

• The developer role is evolving.
  Here’s how to stay ahead. AI is changing how software gets built. Explore the skills you need to keep up and stand out. [LINK]
  

• Codex is now generally available
  We’re announcing the general availability of Codex and three new features that make it even more useful for engineering teams [LINK]

  ---

🔮 What else

• Understanding the 4 main approaches to LLM evaluation (from scratch)
  How do we actually evaluate LLMs? It’s a simple question, but one that tends to open up a much bigger discussion. [LINK]
  

• Every AI founder should be asking these questions

🙌🏼 On Substack

How to AI

Search.

Read more

7 months ago · 87 likes · 8 comments · Ruben Hassid

Hey everyone,

Good morning and happy Tuesday!

+ The news of this week 🗞️ “The AI market is a paradox of soaring valuations and cautionary warnings from leaders like the CEO of Goldman Sachs. This financial tension, however, masks the true narrative: the acceleration of AI adoption is poised to create a seismic shift in the corporate world. As AI becomes more accessible and intelligent, its disruptive force won’t just correct market values but will fundamentally upend industries and displace major players. This post reveals why the true risk isn’t a bubble, but a “train wreck” for those who fail to adapt.” Continue reading here.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

AI Tools

OpenAI has launched AgentKit, a comprehensive suite of tools for developers to build and deploy AI agents, aiming to streamline a process that was previously fragmented and complex. This comes as generative AI continues to reshape the financial services industry, with Google Cloud highlighting how the technology is accelerating innovation in banking and risk management. In a related development, Fujitsu and Sony Bank are partnering to integrate generative AI into their core banking system to speed up development by 20%.

In other news, AI-generated voices have become nearly indistinguishable from real human voices. Politically, President Donald Trump has been using synthetic media to attack opponents and enhance his public image. Additionally, OpenAI is reportedly trying to make ChatGPT a platform, or even a future operating system, with the ability to run apps directly inside the chat window.

In the world of open source, Meta has released OpenZL, a new compression framework, while Tencent’s Hunyuan Image 3.0 has topped the text-to-image rankings.

🦾 AI daily pulse

• [OpenAI] Introducing AgentKit - OpenAI

  Today we’re launching AgentKit, a complete set of tools for developers and enterprises to build, deploy, and optimize agents. Until now, building agents meant juggling fragmented tools—complex orchestration with no versioning, custom connectors, manual eval pipelines, prompt tuning, and weeks of frontend work before launch. [LINK]

• The unstoppable rise of generative AI in financial services

  Gen AI is reshaping financial services, with insights from Google Cloud highlighting how AI is accelerating innovation in banking and risk management. [LINK]
  

• Fujitsu and Sony Bank partner to integrate generative AI into core banking system design and development
  Initiative targets AI-driven system design, 20% faster development, and enhanced service delivery for all core banking functions by April 2026. [LINK]

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• AI voices are now indistinguishable from real human voices

  Do you think you’d be able to tell the difference between a real human voice and a deepfake? Most people can’t. [LINK]
  

• On Truth Social, Trump has embraced AI media to attack foes and boost his image
  The president has posted dozens of pieces of synthetic media, including deepfakes, since he returned to the White House in January, according to an NBC News review. [LINK]

💻 Top techies

• [Meta] Introducing OpenZL: An Open Source format-aware compression framework [LINK]
  

• Tencent’s open-source Hunyuan Image 3.0 hits No.1 on LMArena’s text-to-image board via blind-vote battles; ranking is preliminary and may shift. [LINK]

  ---

🔮 What else

• OpenAI wants ChatGPT to be your future Operating System
  At OpenAI’s Developer Day, CEO Sam Altman showed off apps that run entirely inside the chat window—a new effort to turn ChatGPT into a platform. [LINK]
  

• From Vibe Coding to Vibe Researching: OpenAI’s Mark Chen and Jakub Pachocki

🙌🏼 On Substack

Ahead of AI

Understanding the 4 Main Approaches to LLM Evaluation (From Scratch)

How do we actually evaluate LLMs…

Read more

7 months ago · 215 likes · 15 comments · Sebastian Raschka, PhD

Hey everyone,

Good morning and happy Monday!

This week’s important news 🗞️ “The AI market is a paradox of soaring valuations and cautionary warnings from leaders like the CEO of Goldman Sachs. This financial tension, however, masks the true narrative: the acceleration of AI adoption is poised to create a seismic shift in the corporate world. As AI becomes more accessible and intelligent, its disruptive force won’t just correct market values but will fundamentally upend industries and displace major players. This post reveals why the true risk isn’t a bubble, but a “train wreck” for those who fail to adapt.” Continue reading below.

As a free subscriber, you get the preview from the Premium edition.
Become a paying subscriber and get the most out of the Daily Wild.

Accenture has reported $1.8 billion in new generative AI bookings for the last quarter, while also announcing layoffs for 11,000 employees, it says cannot be “upskilled to AI.” A new study has found that approximately one-quarter of press releases are now written with the help of generative AI.

The week saw significant developments in AI video generation, with both Meta and OpenAI launching new products. Meta introduced Vibes, an AI video feed built in partnership with MidJourney, while OpenAI unveiled Sora 2. In other major news, Google has continued to expand Gemini, adding it to Chrome and Google TV. Additionally, Huawei has released a new open-source technique to reduce the size of LLMs, enabling them to run on less powerful hardware without compromising quality.

Other developments include the launch of a new ethics benchmark for assessing LLMs, details on MetaMask’s airdrop points, and a McKinsey report on the role of CEOs in the agentic age.

🦾 AI daily pulse

• Accenture reported $1.8bn in new ‘generative AI’ bookings in the last quarter
  It will also lay off 11k people it thinks can’t be ‘upskilled to AI’, which got a lot of attention, but it’s also only 1.4% of headcount and might just be a cover. [LINK] [LAYOFFS]
  

• The widespread adoption of large language model-assisted writing across society
  According to this study, based on the analysis of more than 500,000 English-language documents, approximately one-quarter of press releases were written with the assistance of generative AI. Fortunately, writing the CPs is only a small part of the job. [LINK]
  

• This was a big week for AI video generation, with Meta launching Vibes on September 25 and OpenAI following five days later with Sora 2

  Meta moved first with Vibes, an AI video feed within the Meta AI app, built in partnership with MidJourney and Black Forest Labs. The app enables native image-to-video generation with a restyle option and music addition. The app also links to Instagram by allowing remixes when you see a “Vibes” video on your feed. [LINK]

Subscribe to The Daily Wild's Newsletter to unlock the full experience.

⚡️ Top trends

• [Google] Last week, Google added Gemini to Chrome, and this week it’s adding it to Google TV. The launch video does a pretty good job of showing realistic use cases. [LINK]

• Huawei’s new open source technique shrinks LLMs to make them run on less powerful, less expensive hardware
  Huawei’s Computing Systems Lab in Zurich has introduced a new open-source quantization method for large language models (LLMs) aimed at reducing memory demands without sacrificing output quality. [LINK]

💻 Top techies

• MetaMask GitHub releases airdrop points details: Spot trading has the highest weight, and using LINEA points will double your points [LINK]
  

• LLM ethics benchmark: a three-dimensional assessment system for evaluating moral reasoning in large language models. [LINK]

🔮 What else

• [McKinsey Quarterly] The change agent: Goals, decisions, and implications for CEOs in the agentic age [LINK]
  

• AI is upending higher education. Is that a bad thing?

Subscribe now

🗞️ In the news, “Navigating the great AI divide: a ticking clock for the Fortune 500”

The Unseen Risk: Why a market correction isn’t a real threat? Is a corporate mass extinction on the horizon?

Summary: This week, the AI sector presents a tale of two realities. While public markets celebrate record valuations, a growing number of financial leaders, including the CEO of Goldman Sachs, are sounding the alarm on a looming market drawdown, likening the current exuberance to the dot-com bubble. This financial tension, however, distracts from a far more critical narrative: the accelerating pace of AI adoption is poised to create a seismic shift in the corporate world. As AI becomes more accessible and intelligent through its multimodal capabilities, its disruptive force will not only correct market values but will also fundamentally upend industries and displace major players. This post dissects the financial warnings and technological trends to reveal why the true risk isn’t an investment bubble, but a “train wreck” for those who fail to adapt.

Read more