Another Crazy Day in AI 68/110

Another Crazy Day in AI: What Your Face Says About Your Future

Jan 21, 2025
4 min read

Hello, AI Enthusiasts.

As we wrap up the day, let’s uncover what’s happening in the realm of AI! 🌙

Researchers from elite universities have found that AI can predict personality traits from a single facial image, possibly indicating future career and educational success. 🏆

In another study, a recent white paper outlines strategies for integrating security at every phase of the AI lifecycle. Meanwhile, Anthropic CEO believes AI could exceed human capabilities in most tasks within two to three years. 🚀

Here's another crazy day in AI:

How facial recognition might shape your job prospects
Cybersecurity’s role in scaling AI safely
Anthropic CEO affirms AI is reaching a critical milestone
Some AI tools to try out

TODAY'S FEATURED ITEM: The Face Behind the Career: What AI Can Reveal

Image Credit:Wowza (created with Ideogram)

Have you ever wondered if your profile picture could determine your career path?

A new study suggests that artificial intelligence might just have that capability. In an article by Lucas Mearian, Senior Reporter at Computerworld, researchers from several prestigious universities discovered that AI can predict personality traits from a single facial image, which may, in turn, forecast career and educational success. By analyzing images from LinkedIn and top MBA programs, the researchers linked these traits to the actual employment outcomes of 96,000 graduates. This raises important questions about how we perceive talent and success in the hiring process.

Here are some insights from the study:

Personality Insights: AI can identify traits such as Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism based on facial features, linking them to career success.
Hiring Implications: While AI may improve efficiency in selecting candidates, it also brings forth significant issues regarding fairness and bias in recruitment.
Large-Scale Analysis: The study’s broad dataset allows for a robust examination of the relationship between personality traits and key career outcomes, including salary and job changes.
Ethical Challenges: The findings call attention to the ethical implications of using AI in hiring, particularly around issues of bias and individual rights.
Genetic and Environmental Influences: The research illustrates how both genetics and the environment play roles in shaping personality traits and, subsequently, career paths.

Guenzel, Marius and Kogan, Shimon and Niessner, Marina and Shue, Kelly, AI Personality Extraction from Faces: Labor Market Implications (January 09, 2025).

As we consider the implications of using AI in recruitment, it’s crucial to acknowledge both the opportunities and challenges this technology presents. The ability to predict personality from facial images could revolutionize how companies assess candidates, potentially leading to more successful hires. However, relying solely on AI could also reduce individuals to mere images, ignoring the broader context of their experiences and qualifications. Organizations must remain vigilant about the ethical ramifications, ensuring that AI serves to enhance—not replace—the human aspects of hiring.

Moving forward, the conversation about AI in hiring is just beginning. As this technology evolves, it is essential that its integration into recruitment practices emphasizes fairness and inclusivity. Striking a balance between innovative approaches and ethical standards can create a hiring landscape that values each candidate's unique qualities. Ultimately, this will lead to more diverse and dynamic workplaces, where individual stories and experiences are recognized as critical components of success.

Read the full article here.

Read the paper here.

OTHER INTERESTING AI HIGHLIGHTS:

Cybersecurity’s Role in Scaling AI Safely

/Sadie Creese, University of Oxford, and Akshay Joshi, World Economic Forum

Cybersecurity must be a foundational consideration in the adoption of artificial intelligence (AI), ensuring sensitive data is safeguarded and innovation remains responsible. A recent white paper from the World Economic Forum and Oxford University highlights strategies for embedding security at every stage of the AI lifecycle. With only 37% of organizations evaluating AI systems’ security before deployment, leaders must adopt a risk-based approach, align AI initiatives with business goals, and foster multistakeholder collaboration to mitigate vulnerabilities and drive resilient innovation.

Read more here.

Anthropic CEO Affirms AI is Reaching a Critical Milestone

/CNBC Television

Anthropic CEO Dario Amodei shares his perspective on the rapid advancements in artificial intelligence, stating that AI could surpass human capabilities at most tasks within two to three years. Despite occasional setbacks, he likens AI development to a flowing river that finds ways to overcome obstacles. Amodei notes a significant decrease in uncertainty regarding the achievement of human-level AI systems, reflecting optimism about the trajectory of AI innovation.

Read more here.

Watch it here.

SOME AI TOOLS TO TRY OUT:

Stella by FastTracker - Automate meeting invites, emails, to-do lists, reminders, and notes.
Frames by Runway - AI image model with realism and precise stylistic control.
Minduck Discovery - AI search platform with interactive mind maps.

That’s a wrap on today’s Almost Daily craziness.

Catch us almost every day—almost! 😉

EXCITING NEWS:

The Another Crazy Day in AI newsletter is now on LinkedIn!!!

Leveraging AI for Enhanced Content: As part of our commitment to exploring new technologies, we used AI to help curate and refine our newsletters. This enriches our content and keeps us at the forefront of digital innovation, ensuring you stay informed with the latest trends and developments.

Another Crazy Day in AI: The Tools Redefining Materials Science

Jan 20, 2025
4 min read

Hello, AI Enthusiasts.

Good evening and happy Monday! 🌙 As we kick off the work week, let’s highlight some remarkable developments in the world of AI.

The AI for Science team at Microsoft Research has developed two innovative tools that are revolutionizing how we discover and create new materials. 🧪

In other news, we’re on the brink of a significant breakthrough in AI technology, potentially leading to Ph.D.-level super-agents equipped to handle complex human tasks. Meanwhile, a new study shows that GPT-4 has surpassed humans in inverted and displaced Turing tests. 📊

Here's another crazy day in AI:

The story behind MatterGen and MatterSim
Ph.D.-level AI super-agents on the horizon
Turing test results show GPT-4’s human-like precision
Some AI tools to try out

TODAY'S FEATURED ITEM: Dreaming Up the Future of Materials Science

Image Credit:Wowza (created with Ideogram)

What if we could dream up new materials as easily as we sketch ideas on paper?

Microsoft Research's AI for Science team has developed two remarkable tools that are changing how we discover and create new materials. The story begins with Tian Xie, who during his doctoral studies at MIT, pondered whether AI could help generate viable materials based on specific criteria. This question led him to Microsoft Research in 2022, where he joined the AI for Science initiative under scientist Chris Bishop. Today, this vision has evolved into MatterGen and MatterSim, tools that are reshaping our approach to materials discovery.

Here's what makes this significant:

Traditional materials discovery has been like searching through an endless library. MatterGen flips this approach by creating new materials from scratch, similar to an architect designing multiple buildings that meet specific requirements.
MatterSim serves as the reality check, analyzing these new materials to determine which ones could actually exist. This partnership helps researchers validate concepts more efficiently than traditional lab testing.
The project now spans five time zones, bringing together researchers who share a common goal of expanding what's possible in materials science. This diversity in expertise has proven valuable for innovation.
From improving energy storage to advancing medical technology, the practical applications touch multiple fields. The tools are helping researchers explore solutions to real-world challenges.
While receiving recognition in Nature and widespread adoption in the scientific community, the focus remains on practical impact rather than academic achievements.

Materials form the backbone of innovation, enabling everything from the devices we use daily to advancements in medicine and renewable energy. Yet, discovering new materials has often been a slow, resource-heavy process. Tools like MatterGen and MatterSim aim to change that, giving scientists the ability to experiment and refine ideas at unprecedented speeds.

With these AI-driven platforms, researchers can focus on imagining solutions to some of the world’s biggest challenges. Whether it’s improving energy efficiency, creating sustainable materials, or advancing healthcare, the possibilities expand as the barriers to innovation shrink. This collaboration between AI and science demonstrates how technology can enhance human potential, helping us tackle problems faster and more effectively.

What makes this development particularly promising is the sense of shared purpose it fosters. These tools aren’t just about solving individual problems—they encourage partnerships between researchers, industries, and communities. They represent a shift in how we approach innovation, making it more accessible and impactful. With AI at our side, we have the tools to create a future shaped not by limitations, but by our collective ingenuity.

Read the full article here.

OTHER INTERESTING AI HIGHLIGHTS:

Ph.D.-Level AI Super-Agents on the Horizon

/Mike Allen and Jim VandeHei on Axios

A significant breakthrough in AI technology may soon bring Ph.D.-level super-agents capable of tackling complex human tasks. These AI tools could synthesize vast data, analyze options, and deliver high-level results—whether building software, conducting financial analyses, or managing logistics. While advancements promise revolutionary applications in science, health, and education, reliability issues like hallucinations remain a challenge. OpenAI is expected to announce this milestone soon, highlighting AI's potential to reshape industries.

Read more here.

Turing Test Results Show GPT-4’s Human-Like Precision

/Ishika Rathi, Sydney Taylor, Benjamin K. Bergen, and Cameron R. Jones, UC San Diego

A new study from UC San Diego reveals that GPT-4 outperformed humans in inverted and displaced Turing tests, with both human and AI judges often mistaking it for a human participant. The study highlights challenges in distinguishing AI from human interactions in online conversations, underscoring the need for better detection tools. As AI continues to mimic human communication with increasing precision, its potential for deception raises concerns about authenticity in digital spaces.

Read the paper here.

SOME AI TOOLS TO TRY OUT:

Prompt Panda - Manage and use saved prompts directly in AI models.
Pentory AI - Get daily summaries from curated channels.
ReplyPilot - Streamline social media replies while staying on-brand.

That’s a wrap on today’s Almost Daily craziness.

Catch us almost every day—almost! 😉

EXCITING NEWS:

The Another Crazy Day in AI newsletter is now on LinkedIn!!!

Another Crazy Day in AI: A Guide to Sora’s Text-to-Video Magic

Jan 17, 2025
4 min read

Hello, AI Enthusiasts.

It’s Friday night—how’s your week wrapping up? 🌌 Before we unwind for the weekend, here are some intriguing AI updates:

A CNET contributor shared her hands-on experience with Sora, OpenAI’s text-to-video platform that’s transforming video creation and making it more accessible than ever. 🎥

In finance, Goldman Sachs’ CEO revealed that AI can now draft 95% of an IPO prospectus in minutes, significantly reducing the time and manpower required for such tasks. 📄 On the scientific frontier, OpenAI has also entered longevity science, designing proteins for stem cell research in collaboration with Retro Biosciences. 🧬

Here's another crazy day in AI:

A hands-on look at Sora for video creation
AI can draft IPO documents in minutes, says Goldman CEO
OpenAI develops protein engineering AI model for longevity
Some AI tools to try out

TODAY'S FEATURED ITEM: Sora in Action: Journey in Text-to-Video Creation

Image Credit:Wowza (created with Ideogram)

Have you ever imagined turning a simple idea into a high-quality video without the need for a production team or fancy equipment?

Carly Quellman, a contributor to CNET, shares her experience experimenting with Sora, a text-to-video tool from OpenAI. Designed for content creators, marketers, and educators, Sora transforms text prompts into short, photorealistic videos, making video production more accessible for everyone. The platform is accessible through ChatGPT Plus and Pro subscriptions. Carly delves into the platform’s features, usability, and creative possibilities.

What you’ll discover...

What is Sora?
- A new feature for ChatGPT Pro and Plus users that generates short videos or GIFs based on text prompts. It’s aimed at making professional-looking content accessible to everyone.
How it Works
- Input a simple text prompt or use the Storyboard feature to design your vision.
- Choose a style and tone for your video.
- Preview, tweak, and finalize using editing tools like “Remix” or “Storyboard.”
- Export your video or GIF in a format ready to share.
Carly’s Experience
- The AI-generated visuals were surprisingly cinematic and detailed.
- The Storyboard feature offered flexibility, making it easier to refine her ideas.
- While straightforward to use, success often depended on how well she crafted her prompts.
Things to Keep in Mind
- The tool is subscription-based and not yet available in all regions (e.g., EU, UK, Switzerland).
- Overuse without creative input may result in generic outputs.
- Ethical concerns, such as misuse for deepfakes, highlight the need for responsible use.

Tools like Sora highlight how far we’ve come in democratizing creative technology. What once required a dedicated team or expensive equipment can now be done by anyone with an idea and access to AI. For educators, small business owners, or content creators, it offers an exciting way to explore storytelling and engage audiences.

That said, it’s worth remembering that AI tools are just that—tools. Their real power lies in how they amplify human creativity, not replace it. Sora is an impressive companion for turning ideas into visual content, but it’s not a shortcut for meaningful communication. Putting thought into prompts and using the editing features thoughtfully can make the difference between something that feels generic and something that truly connects.

This innovation invites a broader conversation about how we approach AI in creative spaces. Beyond the convenience, there’s an opportunity to think critically about the ethics of its use, ensuring these tools are wielded responsibly. Sora may not redefine video production as a whole, but it offers an intriguing glimpse into what’s possible when AI and creativity intersect.

Read the full article here.

OTHER INTERESTING AI HIGHLIGHTS:

AI Can Draft IPO Documents in Minutes, Says Goldman CEO

/Paolo Confino on Fortune

Goldman Sachs CEO David Solomon revealed that AI could draft 95% of an IPO prospectus in just minutes, significantly reducing the time and manpower required for such tasks. Historically, teams of bankers would take weeks to draft these filings. The efficiency gains exemplify how AI is reshaping investment banking, with institutions like Goldman, JPMorgan, and Wells Fargo integrating AI into operations for productivity, compliance, and client services. As AI adoption grows, banks are increasingly rethinking their workforce composition and client offerings.

Read more here.

OpenAI Develops Protein Engineering AI Model for Longevity

/Antonio Regalado on MIT Technology Review

OpenAI has developed an AI model designed to engineer proteins for longevity science, collaborating with Retro Biosciences on stem cell research. The model, GPT-4b micro, re-engineers proteins to enhance their function, showing improvements over traditional methods in creating Yamanaka factors. These factors can transform skin cells into stem cells, unlocking possibilities for regenerative medicine. While the results are promising, further validation is required to confirm the findings, as OpenAI explores AI’s potential for scientific discovery and artificial general intelligence (AGI).

Read more here.

SOME AI TOOLS TO TRY OUT:

DryMerge - Automate tasks across apps in minutes using plain English.
Ray2 by Luma Labs - AI video model with advanced realism and motion.
Persana AI - AI voice agents and tools to boost sales and marketing.

That’s a wrap on today’s Almost Daily craziness.

Catch us almost every day—almost! 😉

EXCITING NEWS:

The Another Crazy Day in AI newsletter is now on LinkedIn!!!