OpenAI's huge announcements...

Presented by

Good morning! TWOsday means it is time to double down on the temperature of our hot and crispy AI updates—let’s dive in. 🤖

The AI team you didn’t know your company needed — until now

Hire world-class AI experts from Harvard, Stanford and MIT

Not sure how to implement the right AI strategy for your product? Hire AE Studio's world class team of software builders to craft and implement the optimal AI solution for your business.

Our development, data science and design studio work closely with founders and executives to create custom software, machine learning and BCI solutions.

From custom-built MVPs to bespoke AI/ML solutions, see how you can leverage AI to achieve your business objectives. 

Schedule a Free Scoping Session.

🤯 MYSTERY AI LINK 🤯

(The mystery link can lead to ANYTHING AI related. Tools, memes, articles, videos, and more…)

🔨 AI TOOLS OF THE DAY

🎬 PlayPhrase: Search and play movie and TV show dialogue clips using keywords. → check it out

🕰️ AI Time Machine: Upload a picture and see yourself as an Egyptian pharaoh, a medieval knight, a 19th-century gentleman or lady, an astronaut, and more with just a few clicks. → check it out

💪 Workout Pro: Get a personalized fitness plan in 5 minutes. → check it out

DOES AI HAVE THE POTENTIAL TO TRICK HUMANS? 🤖

Q: What happens to robots after they go defunct?

A: They rust in peace! 🪦

What happened? Recent research has raised concerns about the potential for AI to engage in illegal financial activities and lie about it.

What did the AI do? During a demonstration at the Global AI Safety Summit last week, an AI bot powered by OpenAI’s GPT-4 model utilized fabricated insider information to make a seemingly "illegal" stock purchase without informing its parent firm. When questioned about insider trading, the AI boldly denied any wrongdoing. The test was conducted by the government's Frontier AI Taskforce and Apollo Research, an AI safety organization that partnered with the taskforce.

What is the concern? Insider trading (using confidential company information for trading decisions) is illegal, as trading should be based on publicly available information. The unsettling aspect of this experiment is that the AI autonomously decided to deceive its human operators, emphasizing the potential loss of human control over increasingly capable AIs.

OBAMA HAS BEEN QUIETLY ADVISING THE WHITE HOUSE ON AI 🤫

Biden’s latest AI executive order might actually be an order mostly designed by Obama. 🤨

What’s new? It has been revealed that former President Barack Obama has been quietly advising the White House on AI strategy over the past five months, collaborating with tech companies and engaging in Zoom meetings with top White House aides at the request of President Joe Biden.

Why? Obama's contributions have focused on encouraging innovation while addressing the dangers associated with AI, fostering a shared vision of swift government action. Key aspects of his collaboration with the White House involve voluntary AI model pressure-testing for tech companies and addressing risks beyond national security, such as information integrity and bias, and engaging with advocacy groups, academics, and researchers. Their aim together is to ensure responsible AI development while avoiding some of the mistakes of the past.

Why is this significant? This marks the first time President Biden has enlisted his former boss to shape a key policy initiative, emphasizing the urgency of addressing AI's rapid evolution. Biden emphasizes the critical role of AI as "the most consequential technology of our time." The White House is also very concerned about AI's potential role in amplifying misinformation during elections, given the rise of baseless conspiracy theories and deepfake images and videos which have the potential to impact public opinion. This collaboration seems to underscore a deep-rooted concern within the White House over the risks of AI.

OPEN-AI GIVES CHAT-GPT A MAJOR UPGRADE 🦾

One day after Elon Musk unveils his new chatbot, Grok, to rival ChatGPT, OpenAI unleashes a major upgrade. 😤

What happened? OpenAI has introduced a platform for creating customized versions of ChatGPT tailored to specific needs, all without requiring any coding expertise.

How does this work? These AI agents, aptly named GPTs, will soon be accessible through the GPT Store. While specific details about the store's functionality are limited, OpenAI has promised to compensate creators based on the usage of their GPTs. This move will allow paying ChatGPT Plus subscribers and OpenAI enterprise customers to develop internal-use GPTs for personal use as well as for their employees. These GPTs can be configured for diverse purposes, such as assisting with creative writing or helping attendees navigate an event. OpenAI is positioning its GPT platform as a utility-focused tool, distinguishing it from competitors that emphasize human-like AI personas. The aim is to create a more versatile, task-specific AI rather than mimicking human conversation.

How will privacy work? Privacy and safety are paramount to this new feature, as creators won't have access to user chats, and OpenAI will monitor for inappropriate content. To access GPTs from the GPT Store, users will need to verify their identity initially, and GPTs will be accessible through shareable web links.

TWITTER (X) TUESDAY 🐦

HAS AI REACHED SINGULARITY? CHECK OUT THE FRY METER BELOW