• The AI Pulse
  • Posts
  • 🤖 OpenAI CEO Sam Altman Teases “Strawberry”

🤖 OpenAI CEO Sam Altman Teases “Strawberry”

PLUS: Anthropic Offers $15,000 Bounties to Hackers, AI Accelerates Schizophrenia Cure

Author

James Roesser
August 09, 2024

Welcome back AI enthusiasts!

In today’s AI Report:

  • 🍓OpenAI CEO Sam Altman Teases “Strawberry”

  • 💵Anthropic Offers $15,000 Bounties to Hackers

  • 🔬AI Accelerates Schizophrenia Cure

  • 🛠5 Trending Tools

  • 💰Venture Capital Updates

  • 💼Who’s Hiring?

Read Time: 3 minutes

🗞RECENT NEWS

OPENAI

🍓OpenAI CEO Sam Altman Teases “Strawberry”

Image Source: Canva AI Image Generator

Mysterious AI models called “anonymous-chatbot” and “sus-column-r” appeared in the LMSYS Chatbot Arena.

Key Details:

  • The LMSYS Chatbot Arena is an open-source platform that evaluates AI models across various criteria, including the Massive Multitask Language Understanding (MMLU) benchmark, which covers 16,000 questions from 57 academic disciplines.

  • The mysterious AI models were available in the LMSYS Chatbot Arena’s “Battle” section, which allows website visitors to compare how AI models tackle user queries to generate outputs.

  • Many believe “anonymous-chatbot” is OpenAI’s new AI model called “Strawberry,” which scans the internet autonomously to solve more complex real-world problems at scale.

  • To fuel the speculation, OpenAI CEO Sam Altman posted a strawberry image on X with the caption: “I love summer in the garden.”

  • OpenAI has previously tested unreleased AI models in the LMSYS Chatbot Arena, with “gpt2-chatbot” appearing before releasing GPT-4o (“o” for “omni”).

Why It’s Important:

  • Q* (“Q-Star”) refers to OpenAI’s project to develop an AI algorithm that automates processes and solves some of the world’s most pressing issues.

  • OpenAI users speculate “anonymous-chatbot” creates the conditions for Q* (“Q-Star”) to become a reality.

🩺 PULSE CHECK

Is “anonymous-chatbot” OpenAI’s new AI model called “Strawberry”?

Vote Below to View Live Results

Login or Subscribe to participate in polls.

ANTHROPIC

💵Anthropic Offers $15,000 Bounties to Hackers

Image Source: Kimberly White/Getty Images for TechCrunch/Flickr

Anthropic, an AI research company that builds reliable AI systems, launched a bug bounty program to identify critical vulnerabilities in AI systems.

Key Details:

  • The program focuses on identifying and mitigating “universal jailbreak attacks”: a method to circumvent an AI system’s built-in safety measures and ethical guidelines.

  • Anthropic offers bounty rewards of up to $15,000 for identifying vulnerabilities in the Chemical, Biological, Radiological, and Nuclear (CBRN) domains.

  • Anthropic aims to prevent bad actors from manipulating AI systems to elicit harmful responses that spread misinformation in critical, high-risk CBRN domains.

Why It’s Important:

  • Finding weaknesses in AI systems leads to more robust and secure AI models, which in turn helps foster public trust in AI technology.

  • Discovering novel ways to manipulate AI systems can lead to groundbreaking research in safety, security, and privacy measures.

🚨If you’re experienced in identifying jailbreaks in AI models, fill out this application form by August 16th at 11:59 PM (PDT).

AI RESEARCH

🔬AI Accelerates Schizophrenia Cure

Researchers at Uppsala University (UU) developed an AI model to create three-dimensional (3D) structures of receptors linked to schizophrenia.

The AI model generated Trace Amine-Associated Receptor 1 (TAAR1) structures, a protein molecule that detects chemicals linked to mental health.

Then, supercomputers searched chemical libraries containing millions of molecules to find those fitting the AI-generated TAAR1 structures.

“The accuracy of the AI-generated TAAR1 structures was astonishing—I couldn’t believe it,” said Jens Carlsson, a professor in Computational Biochemistry at the Department of Cell and Molecular Biology.

🛠TRENDING TOOLS

📑Tezi is an autonomous recruiting partner.

🧸BrainyBear builds AI chatbots in seconds.

📊Flowith streamlines tasks on a multi-thread interface.

🔁ForLoop accelerates website scraping and data automation.

🎓SemanticScholar is a free AI-powered research tool for scientific literature.

🔮Browse our always Up-To-Date AI Tools Database.

💰VENTURE CAPITAL UPDATES

  • Assisterr secures a $1.7M Pre-Seed to scale tokenized AI models.

  • Masterplace scores a $2M Seed Round to launch a first-of-its-kind AI ed-tech platform.

  • LevelAI secures a $39.4M Series C for customer experience innovation.

💼WHO’S HIRING?

  • Tesla (Palo Alto, CA): Authentication Embedded Developer, Vehicle Firmware, Spring 2025

  • QuantiqPartners (Austin, TX): Software Developer Intern, Summer 2025

  • Uber (New York, NY): Software Engineering UberSTAR Intern, Summer 2025

  • NVIDIA (Santa Clara, CA): Software Engineer, VLSI Design Tools, New Grad 20204

  • Roblox (San Mateo, CA): Software Engineer, New Grad 2025

🤖PROMPT OF THE DAY

E-COMMERCE

👷‍♂️Create E-Commerce Platform

Guide me through creating an E-Commerce Platform for [Business] with [Product/Service] in [Industry]. Explain how to register a domain name, select a website hosting provider, and design an optimized website layout.

Business = [Insert Here]

Product/Service = [Insert Here]

Industry = [Insert Here]

📒FINAL NOTE

If you found this useful, follow us on Twitter or provide honest feedback below. It helps us improve our content.

How was today’s newsletter?

❤️TAIP Review of the Day

“I LOVE the consistent, balanced, and neutral summaries of the latest AI developments. It’s all facts, no feelings.”

-Logan (⭐️⭐️⭐️⭐️⭐️Nailed it!)
REFER & EARN

🎉Your Friends Learn, You Earn!

You currently have 0 referrals, only 1 away from receiving ⚙️Ultimate Prompt Engineering Guide.

Refer 5 friends to enter 🎰July’s $200 Gift Card Giveaway.

Reply

or to participate.