Another Crazy Day in AI 64/110

Another Crazy Day in AI: Associations are Rethinking their Marketing Playbook

Feb 6, 2025
3 min read

Hello, AI Enthusiasts.

Thursday night wind-down—AI never sleeps, but we can try! 😴

Marketing General Incorporated (MGI) reveals that associations are optimistic about 2025—but they’re revamping marketing strategies and investing in tech to keep up.

Meanwhile, Google’s new AI policy framework lays out a vision for scientific breakthroughs, pushing for AI-driven research. And in the world of robotics, AI-powered humanoid robots are taking movement to the next level—dancing, sparring, and mimicking human gestures with eerie precision. 🤖💃

Here's another crazy day in AI:

Why associations are doubling down on marketing tech
Google’s vision for AI-driven scientific discovery
AI-powered robots are learning to dance
Some AI tools to try out

TODAY'S FEATURED ITEM: What’s Changing in Association Marketing

Image Credit:Wowza (created with Ideogram)

Is email marketing losing its edge? Associations think so...

Association professionals are heading into 2025 with optimism about membership growth and revenue—but they’re also realizing they need to level up their marketing strategies. A new report from Marketing General Incorporated (MGI), featured by Mark Athitakis in Associations Now, reveals that associations are planning to invest more in technology and diversify their outreach beyond traditional email.

Source: Associations Now/Marketing General Incorporated

What’s shaping association marketing in the year ahead?

Bigger investments in marketing and technology – More than half of respondents plan to increase spending on member marketing (56%) and new technology (53%).
Adapting to changes in email marketing – With stricter spam filters from email providers, 37% of associations are exploring alternative communication channels.
Cautious approach to AI – Only 3% of associations currently use AI for member services, and just 19% plan to in 2025, primarily for content creation rather than analytics.
Greater focus on exclusive content – 43% of respondents plan to emphasize publications and specialized resources, a significant increase from the previous year.

Source: Marketing General Incorporated, 2025 Association Outlook Report — Source: Marketing General Incorporated, 2025 *Association Outlook Report*

For years, email has been the go-to communication tool for associations, but changing algorithms and inbox filtering are making it harder to rely on alone. That’s why many organizations are beginning to experiment with alternative outreach methods like digital advertising, retargeting, and even ringless voicemail. The shift isn’t just about technology—it’s about making sure members actually see and engage with the content that’s meant for them.

At the same time, associations are recognizing that access to unique, high-value content is a major driver of membership. With so much information freely available online, organizations are looking for ways to offer something members can’t easily find elsewhere. The challenge moving forward will be figuring out how to blend technology with personalization—leveraging new tools while maintaining the trust and connection that keep members engaged.

Read the full article here.

Download a copy of the report here.

OTHER INTERESTING AI HIGHLIGHTS:

Google’s Vision for AI-Driven Scientific Discovery

/Kent Walker, President of Global Affairs, Google & Alphabet, and James Manyika, SVP, Research, Labs, Technology & Society

AI is transforming scientific research, accelerating discoveries in fields like biology, climate science, and medicine at an unprecedented pace. Google’s new Policy Framework for Building the Future of Science with AI outlines key steps for policymakers to ensure continued AI-driven progress. The framework emphasizes three core pillars: Infrastructure (expanding AI resources for researchers), Investment (funding AI-powered scientific innovation), and Innovation (pro-innovation policies that remove legal barriers to progress). As AI’s role in science expands, global collaboration and supportive regulations will be essential for unlocking its full potential.

Read more here.

AI-Powered Robots Are Learning to Dance

/Rory Bathgate, LiveScience

A new AI system called ExBody2 allows humanoid robots to move with remarkable human-like fluidity, even performing dances, sparring sessions, and complex gestures with realistic motion. Developed by researchers from UC San Diego, MIT, UC Berkeley, and Nvidia, ExBody2 uses motion-capture data and reinforcement learning to teach robots fine motor skills and adaptive movement. The breakthrough could lead to robots that navigate dynamic environments more naturally, improving their use in fields like healthcare, retail, and manufacturing. The next challenge is training robots to perform these tasks without manually curated datasets, further automating their learning process.

Read more here.

SOME AI TOOLS TO TRY OUT:

Riffo - AI-powered file naming that matches content.
Jace - AI email agent that drafts replies, schedules meetings, and researches context.
Pressdeck - Instantly create a searchable press website, replacing PDF kits and folders.

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: CSU brings ChatGPT to Thousands of Students and Faculty

Feb 5, 2025
3 min read

Hello, AI Enthusiasts.

The week’s in full swing! How’s it treating you?

The California State University (CSU) system is deploying ChatGPT Edu across 23 campuses, providing AI-powered learning to over 460,000 students and 63,000 faculty with an initiative to enhance education and workforce readiness. 📚

Elsewhere, Google has rolled out expanded versions of Gemini 2.0, fine-tuned for different applications. And if that’s not enough, ByteDance’s new deepfake system, OmniHuman-1, can generate ultra-realistic AI videos from a single image and audio sample. 😲

Here's another crazy day in AI:

The largest AI rollout in higher ed is here
Google rolls out Gemini 2.0 to all users
AI deepfakes just got more real than ever
Some AI tools to try out

TODAY'S FEATURED ITEM: ChatGPT Makes Its Way Into The CSU System

Image Credit:Wowza (created with Ideogram)

What if AI was as essential to college as Wi-Fi?

The California State University (CSU) system is making history by rolling out ChatGPT Edu to over 460,000 students and 63,000 faculty and staff across 23 campuses. This marks the largest deployment of ChatGPT in education, positioning CSU as a leader in AI-powered learning. OpenAI, the organization behind this initiative, sees AI literacy as an essential skill for the future workforce and is working to make AI tools more accessible in higher education.

What this means for students and faculty

Smarter learning tools – Students will have access to AI-powered tutoring, study aids, and interactive course-specific GPTs.
More efficient teaching – Faculty can use AI to assist with lesson planning, curriculum development, and administrative work.
AI training and certifications – CSU is launching free AI education programs to help students and staff develop essential AI skills.
Stronger career opportunities – Students will be connected with AI-focused apprenticeships to build real-world experience.

With more industries adopting AI, universities are under pressure to ensure students graduate with the skills needed to keep up. CSU’s approach isn’t just about making AI accessible—it’s about integrating it into education in a way that enhances learning while maintaining academic integrity. By giving students and faculty access to AI tools at scale, CSU is making a statement: AI isn’t just something to study; it’s something to use, responsibly and effectively.

This shift also raises important questions. How will AI impact traditional learning methods? What are the risks of over-reliance on AI-generated content? And how can universities ensure that students develop critical thinking skills alongside technological proficiency? While AI can be a powerful tool, its real impact will depend on how it is used. CSU’s initiative provides a model for other institutions, but it will be the students and educators themselves who ultimately shape what AI in education looks like.

Read the full article here.

OTHER INTERESTING AI HIGHLIGHTS:

Google Rolls Out Gemini 2.0 to All Users

/Koray Kavukcuoglu, CTO, Google DeepMind, on behalf of the Gemini team

Google has expanded the availability of Gemini 2.0, introducing multiple versions tailored for different use cases. The new Gemini 2.0 Flash is optimized for speed and efficiency, while Gemini 2.0 Pro delivers enhanced coding and complex reasoning capabilities. The introduction of Gemini 2.0 Flash-Lite makes AI more cost-efficient, providing strong performance at a lower price. With multimodal input, massive context windows, and reinforced safety measures, the Gemini 2.0 lineup marks a significant step forward in AI accessibility and capability.

Read more here.

AI Deepfakes Just Got More Real Than Ever

/Kyle Wiggers, TechCrunch

ByteDance researchers have unveiled OmniHuman-1, a cutting-edge deepfake system capable of generating eerily realistic AI videos with just a single reference image and audio sample. Unlike previous deepfake models, OmniHuman-1 produces seamless, high-quality videos that bypass the uncanny valley, making AI-generated impersonations even harder to detect. While not yet publicly released, the rapid advancements raise concerns over political misinformation, financial fraud, and identity theft. As deepfake technology improves, calls for stricter regulations and detection methods are growing.

Read more here.

SOME AI TOOLS TO TRY OUT:

Orate - Generate human-like speech and transcribe audio with a unified AI-powered API.
Swatle - AI that breaks down project tasks into manageable subtasks with time estimates.
Earkick - Anonymous AI mental health app with no registration or data tracking.

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: The Psychology Behind Machine Intelligence

Feb 4, 2025
3 min read

Hello, AI Enthusiasts.

As the day wraps up, let’s explore the intriguing connections in the realm of AI! 🌙

While we often think about algorithms and data in AI discussions, the influence of psychology is foundational. Delve into how insights from the human mind have guided the development of AI and continue to shape its future. 🧠

Plus, learn why many employees are using AI tools at work without approval, a practice known as "shadow AI." On another note, Meta has launched its Frontier AI Framework to guide risk assessment and model release decisions.

Here's another crazy day in AI:

The brain science behind AI
Why workers are smuggling AI into their jobs
Meta’s open-source approach to AI safety
Some AI tools to try out

TODAY'S FEATURED ITEM: Before AI, There Was Psychology

Image Credit:Wowza (created with Ideogram)

What if understanding the human mind is the key to building smarter machines?

When we think about artificial intelligence, we often focus on algorithms, data, and computing power. But long before AI became what it is today, psychologists were already laying the groundwork by studying how humans learn, adapt, and solve problems. This article by Chris Ludlow and Armita Zarnegar, featured in The Conversation, explores the deep connection between psychology and AI—how insights from the human brain helped shape machine intelligence and continue to influence its development today.

How the human mind shaped machine intelligence:

Neural networks began with human learning theories – Donald Hebb’s 1949 model of learning inspired early artificial neural networks.
Machines that learn like humans – Frank Rosenblatt’s perceptron, developed in the 1950s, was built on Hebb’s ideas, marking AI’s first steps in pattern recognition.
Deep learning’s psychological foundation – David Rumelhart and colleagues advanced neural networks in the 1980s, leading to the deep learning models we use today.
The challenge of self-awareness – AI lacks metacognition, or the ability to reflect on its own thinking, a concept psychologist John Flavell explored in human cognition.
Problem-solving without prior experience – AI researchers are turning to psychological theories to help machines make decisions in unfamiliar situations.
The limits of AI reasoning – Studies by Daniel Kahneman suggest that AI explanations might be just as flawed as human reasoning biases.
AI’s impact on human intelligence – Just as taxi drivers’ brains adapt to navigation, AI could be shaping how we process information and solve problems.

Psychology has influenced AI for decades, providing insights into learning, decision-making, and adaptation. But the relationship goes both ways. As AI continues to advance, it’s also challenging our understanding of human intelligence. What does it mean to "think" or "learn" when a machine can process vast amounts of information in ways we can't?

At the same time, AI is changing how we interact with the world. From recommendation systems to automated decision-making, it’s influencing our choices and perceptions, sometimes in ways we don’t fully understand. As this technology continues to evolve, psychology will remain essential—not just for improving AI, but for helping us understand its impact on human behavior and society.

Read the full article here.

OTHER INTERESTING AI HIGHLIGHTS:

Why Workers Are Smuggling AI Into Their Jobs

/Sean McManus, Technology Reporter on BBC

Many employees are using AI tools at work without approval, a practice known as "shadow AI." Workers cite reasons like IT restrictions, slow approvals, or a preference for better tools. While AI boosts productivity, it also raises security concerns, as unauthorized AI use could expose sensitive data. Companies are now grappling with whether to ban AI or find ways to safely integrate it into workflows.

Read more here.

Meta’s Open-Source Approach to AI Safety

/Meta

Meta has introduced its Frontier AI Framework to guide risk assessment and model release decisions, aligning with global AI safety commitments. The framework focuses on cybersecurity threats and risks related to chemical and biological weapons, balancing innovation with security. By maintaining an open-source approach, Meta aims to encourage collaboration while mitigating catastrophic risks. As AI evolves, Meta plans to refine its strategy to ensure responsible development.

Read more here.

Check the framework here.

SOME AI TOOLS TO TRY OUT:

Tana - An AI-native workspace that turns notes into tasks, webpages, and more.
Sonofa - Converts reading materials into podcasts for on-the-go listening.
Marketeam - AI-powered assistants for content marketing, SEO, and AEO.

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!!!