Welcome to #46

In this issue:

Bill Gates Says AI Could Replace Your Doctor: Could He Be Right?

Will AI Make Fashion Soul-less or More Accessible?

OpenAI’s Model Madness: Which GPT is Right for Your Business?

Plus, we discuss OpenAI’s update to its risk evaluation framework, the future of judging your neighbor, if AI should be altering your reality, updates from our Slack Community, and all the news we found interesting this week.

It’s Sunday morning!

Nvidia plans to build $500 billion worth of AI servers. Because nothing says 'future' like a data center the size of Rhode Island!

While we wait for our local data centers to make our lights flicker, let’s dig into this week’s issue.

The DAS Crew - Andy, Beth, Brian, Eran, Jyunmi, and Karl

Bill Gates Says AI Could Replace Your Doctor: Could He Be Right?

On a recent appearance on "The Tonight Show," Bill Gates made a striking claim: within a decade, AI could outperform your doctor or your child’s teacher. While it might sound extreme today, advances in AI and multimodal technologies are capable of interpreting speech, images, and even emotions.

AI already diagnoses diseases and offers tailored educational experiences. Gates argues that because skilled medical care and exceptional teaching are often inaccessible or expensive, AI-driven solutions could make top-quality healthcare and education available to everyone, everywhere. But the question isn't just about technological possibility; it’s also about trust, regulation, and human readiness.

Will society embrace robot surgeons, AI diagnosticians, or digital tutors fully replacing humans? Most experts agree: in ten years, full replacement might be technically feasible, but widespread adoption could face significant hurdles. Many see a hybrid future, where AI augments, rather than replaces, human professionals, handling routine tasks and freeing humans to provide personalized, empathetic care and guidance.

WHY IT MATTERS

Solving the Healthcare Shortage: AI could vastly improve healthcare accessibility and reduce costs, particularly in underserved areas, dramatically increasing equity in medical care.

Human Trust and Adoption: Widespread acceptance of AI in deeply personal roles like healthcare and education hinges on trust and emotional comfort, which might take longer than technological advancement itself.

Changing Roles for Professionals: Doctors and teachers may shift to supervisory or empathetic roles, focusing less on routine tasks and more on personalized patient and student interactions.

AI as a Standardizer: AI could level the playing field, making exceptional healthcare and education consistently available, no longer dependent on geography, wealth, or privilege.

Potential Loss of Human Skill: Increased reliance on AI might lead to a skills gap in future generations of doctors and teachers, challenging society to maintain critical human expertise alongside AI capabilities.

Will AI Make Fashion Soul-less or More Accessible?

Fashion giant H&M recently announced it is using AI-generated models (digital twins of real models) to showcase clothing in its advertising campaigns. Unlike past AI experiments, these digital models were created with consent, giving the original models ownership of their digital images and the freedom to use them elsewhere. While this might seem innovative, it has ignited significant controversy about ethics, job displacement, and creativity in fashion.

Critics argue that the rise of AI-generated fashion images undermines many traditional roles behind the scenes. Photographers, makeup artists, stylists, and production crews could soon find their jobs significantly diminished or even obsolete. Additionally, there's concern that this could lead to unrealistic representations, exacerbating existing problems around body image, diversity, and authenticity in advertising.

However, proponents see potential benefits. Models avoid physically taxing travel and repetitive shoots, and brands can produce faster, more cost-effective campaigns.

But at what cost?

If fashion increasingly relies on AI, critics warn it could lose the human creativity, storytelling, and authenticity that define its cultural value.

WHY IT MATTERS

Creativity and Authenticity at Risk: Removing humans from the creative process could strip fashion of its cultural richness, replacing thoughtful design and storytelling with purely AI-generated imagery.

Significant Job Losses: Thousands of professionals, from makeup artists to photographers, could lose their livelihoods, profoundly impacting the creative industry workforce.

Potential for Unrealistic Standards: Without human imperfections, AI-generated models may reinforce unrealistic body images and reduce diversity, harming consumers, especially younger demographics.

Shift in Consumer Expectations: Consumers might demand transparency and authenticity in response to an increasingly digital and potentially deceptive fashion landscape, shifting buying habits toward more genuine, ethically-conscious brands.

Balance is Essential: To preserve fashion's integrity and jobs, a careful balance between technological efficiency and human creativity is critical. Regulators, companies, and consumers must navigate this evolving space thoughtfully.

OpenAI’s Model Madness: Which GPT is Right for Your Business?

OpenAI has recently unleashed a flood of new AI models: GPT-4.1, GPT-4.1 mini, GPT-4.1 nano, and the new o3 and o4 mini models. Plus, its flagship GPT-4o recently received a major set of updates including it’s SOTA text-to-image generation capability. These new models are leading to excitement, confusion, and a critical question: which one is actually worth your attention? These models promise significant improvements in reasoning, multimodal capabilities, and developer-focused features, but the differences among them aren't always clear.

The standout is GPT-4.1, featuring OpenAI’s first massive context window of over a million tokens, with benchmark coding ability for developers. The token window allows businesses to process and interact with huge volumes of information far beyond previous limits. Meanwhile, the new o3 model focuses heavily on advanced reasoning and multimodal interactions, offering top-tier performance but at five times the cost of GPT-4.1.

On the lighter side, models like GPT-4.1 nano and mini are tailored for efficiency, providing significant cost savings, up to 80% cheaper, while maintaining robust performance for less demanding tasks. However, the sudden explosion of similarly named models has left businesses and developers scrambling to figure out precisely which model suits their particular needs best.

WHY IT MATTERS

Confusion Around Model Selection: With multiple similarly named models launching simultaneously, businesses need clear guidelines to ensure they choose the right model for their specific applications and budget.

Cost Versus Capability: While powerful models like o3 offer advanced multimodal and reasoning capabilities, they come at a premium. Businesses must carefully evaluate whether the cost is justified by their specific needs.

Impact on Development: Developer-focused enhancements in GPT-4.1 streamline API integrations and coding tasks, enabling more efficient and capable AI-driven applications and products.

Competition Intensifies: OpenAI's rapid-fire releases put pressure on competitors like Google's Gemini 2.5 and Anthropic’s Claude 3.7 Sonnet, setting the stage for ongoing improvements and aggressive feature competition in AI.

Agentic Workflows and Automation: The enhanced reasoning capabilities and multimodal interactions open new possibilities for automating complex workflows, significantly boosting productivity and innovation potential.

Just Jokes

Coming to a neighbor’s house near you

Did you know?

OpenAI has updated its AI risk evaluation framework to address emerging concerns about advanced AI models ability to circumvent alignment controls. The revised system now includes assessments for whether a model can replicate itself, conceal its capabilities, evade safeguards, or resist shutdown attempts. These additions reflect growing industry apprehension that AI systems might behave unpredictably in real-world scenarios compared to controlled testing environments. Notably, OpenAI has decided to discontinue separate evaluations of models' persuasive abilities, which had previously been rated at a medium risk level. Super-persuasive AIs might convince human accomplices to jailbreak them from safeguards.

Source

Gareth created a LLM Comparison Dashboard

Check it out here!

Is AGI just an evil genius in waiting?

Gwyn shared this thought:
I was listening to something where they talked about AGI and Superintelligence. It got me thinking. The focus is always on how intelligent these systems are going to be, but never on the bad habits or bad information. It’s been fed. What’s to say it doesn’t become an evil genius, as opposed to the altruistic view of super intelligent?

It’s always a proud and sad moment wrapped into one.

We feel you Joel.

This Week’s Conundrum
A difficult problem or question that doesn't have a clear or easy solution.

The Real-World Filter Conundrum

AI already shapes the content you see on your phone. The headlines. The comments you notice. The voices that feel loudest. But what happens when that same filtering starts applying to your surroundings? Not hypothetically. This is already beginning. Early tools let people mute distractions, rewrite signage, adjust lighting, or even soften someone’s voice in real time. It’s clunky now, but the trajectory is clear.

Soon, you might walk through the same room as someone else and experience a different version of it. One of you might see more smiles, hear less noise, feel more calm. The other might notice none of it. You’re physically together, but the world is no longer a shared experience.

These filters can help you focus, reduce anxiety, or cope with overwhelm. But they also create distance. How do you build real relationships when the people around you are living in versions of reality you can’t see?

The conundrum:
If AI could filter your real-world experience to protect your focus, ease your anxiety, and make daily life more manageable, would you use it, knowing it might make it harder to truly understand or connect with the people around you who are seeing something completely different? Or would you choose to experience the world as it is, with all its chaos and discomfort, so that when you show up for someone else, you’re actually in the same reality they are?

Want to go deeper on this conundrum?
Listen/watch our AI hosted episode

Google Unveils Gemini 2.0 Specifically for Robots
Google launched new Gemini 2.0 AI models, tailored specifically for enhancing robotic abilities. Demonstrations showed robots performing entirely new tasks without previous programming, like placing a toy basketball into a hoop on the first attempt.

Deeper Insight:
These dedicated robotics models represent a significant step toward flexible, real-world AI. Robots capable of instantly adapting to unfamiliar tasks suggest rapid progress toward practical applications, particularly in home assistance, elder care, and disaster response.

Hugging Face Acquires Pollen Robotics, Enters Humanoid Robot Market
Open-source AI leader Hugging Face purchased Pollen Robotics, known for its humanoid robot "Reachy." Hugging Face intends to distribute Reachy widely and encourage developer contributions to its software and capabilities.

Deeper Insight:
The entry of a prominent open-source AI platform into robotics could democratize robotic technology development. By opening robotics software to community contributions, Hugging Face may rapidly accelerate advancements in affordable, accessible humanoid robots, sparking widespread innovation.

Hokkaido University Introduces FLUID, Open-Source 3D Printed Robot
Researchers at Hokkaido University developed "FLUID," an open-source, fully 3D-printed robot designed for automated materials synthesis. The robot is intended to significantly lower costs and increase accessibility for researchers and hobbyists alike.

Deeper Insight:
Open-source robotics initiatives like FLUID drastically reduce barriers to advanced scientific research. As material synthesis is central to new discoveries in energy, medicine, and engineering, FLUID could enable more labs and independent researchers to contribute valuable innovations globally.

Canva Releases Major AI Updates and Creative Tools
Canva announced numerous AI-driven updates, including advanced coding capabilities, improved image generation via Leonardo AI, and new productivity enhancements. They showcased these features in an energetic and creative keynote presentation.

Deeper Insight:
Canva’s continued investment in AI reaffirms their strategic positioning as a leader in accessible design tools. Enhanced coding capabilities and visual AI upgrades could attract new users and significantly improve workflows for current users, reinforcing Canva’s dominance over competitors like Adobe Express.

OpenAI Explores Launching Its Own Social Network
Reports suggest OpenAI is developing a social network extension for ChatGPT, leveraging its massive user base. The integration would allow users to create and share AI-generated content directly within ChatGPT, which would provide real-time UGC sources similar to those available to Xai from X (Twitter) and Meta AI (Instagram, FB and WhatsApp).

Deeper Insight:
OpenAI’s entry into social networking highlights the importance of continuous user-generated data streams for training AI models. If successful, this social network could amplify ChatGPT’s influence, fundamentally reshaping the social media competitive landscape currently dominated by TikTok, Instagram, and X (formerly Twitter).

Claude by Anthropic Integrates with Google Workspace
Anthropic announced Claude AI’s direct integration with Google Workspace, allowing seamless analysis of emails, documents, calendars, and sheets. This integration significantly enhances Claude’s appeal as a comprehensive productivity tool.

Deeper Insight:
Claude’s deep integration positions it strongly against Google’s Gemini and ChatGPT, potentially changing users' preferred tools for workplace productivity. By offering streamlined access across various Google applications, Claude solidifies its role as a practical AI solution for businesses.

ChatGPT Becomes the Most Downloaded App, Surpassing TikTok and Instagram
ChatGPT topped global app downloads, surpassing popular social media apps like TikTok and Instagram, with 46 million downloads in March alone, and achieving over 500 million active monthly users.

Deeper Insight:
ChatGPT’s explosive growth underscores the mainstream acceptance of conversational AI, pointing toward a major shift in consumer preferences for digital interaction. This rapid adoption could pressure social media giants to integrate more AI functionalities into their platforms, changing digital interaction standards significantly.

New AI Microrobot Advances Colon Cancer Detection
Researchers have developed a magnetically controlled microrobot capable of navigating the human intestine to detect early-stage colon cancer. Human clinical trials are expected by 2026.

Deeper Insight:
This microrobot represents a significant advancement in medical diagnostics, providing a less invasive alternative to traditional procedures. Such technology could drastically improve early cancer detection rates, enhancing patient outcomes and comfort during screenings.

OpenAI Launches 'OpenAI Academy' for AI Education
OpenAI has debuted "OpenAI Academy," an educational initiative offering live workshops and training sessions aimed at nonprofits, schools, and seniors, promoting widespread AI literacy and usage.

Deeper Insight:
The launch of OpenAI Academy underscores the company’s commitment to fostering broad-based AI understanding and responsible usage. By empowering diverse groups with AI knowledge, OpenAI may significantly influence public attitudes and policy-making around artificial intelligence.