In today’s newsletter:

• News - 🌍 ChatGPT’s AI Search Goes Free for All Users!

• Culture - 🤖 AI Chatbots Show Bias: Less Empathy for Certain Races

🌍 ChatGPT’s AI Search Goes Free for All Users!

ChatGPT’s AI-powered search engine is now available to all users, including those on the free tier, as part of OpenAI’s “12 Days of Ship-mas” announcements. Initially launched for paid subscribers in October, this feature is designed to deliver faster, more interactive search results. On mobile, ChatGPT now mimics a traditional search engine by displaying results with images, ratings, and maps for queries like restaurants or local attractions. Users can click for more details or directions, blending AI’s conversational power with the convenience of real-time search.

For advanced users, ChatGPT's updates include faster link previews for searches like “hotel booking websites” and real-time web updates in Advanced Voice Mode (exclusive to paid plans). These upgrades come alongside OpenAI’s broader innovations, including the launch of its text-to-image model Sora and a premium $200/month ChatGPT Pro subscription. As ChatGPT evolves into a multifunctional platform, it’s clear OpenAI is ramping up efforts to blend AI capabilities with the familiar tools we rely on daily.

🗞️ In Other News…

• Wedbush raised SoundHound AI's price target to $22, citing its innovative voice AI technology, Nvidia partnership, strategic acquisitions, and strong growth potential in the speech AI market, despite valuation pressures.

• AI voice cloning scams are on the rise, with criminals using technology to impersonate loved ones and exploit victims, but experts recommend combating these frauds by implementing family safe words and strict identity verification protocols.

• Brazilian football icon Adriano bid an emotional farewell at Flamengo’s Maracana Stadium, marked by a heartfelt AI message from his late father, showcasing his storied career, personal struggles, and enduring legacy.

• Lockheed Martin launched Astris AI, a subsidiary offering secure, scalable AI solutions and expert consulting services to strengthen the defense industrial base and high-assurance commercial industries.

• Trump’s Silicon Valley advisers, including Elon Musk and Marc Andreessen, are pushing back against AI "censorship," claiming tech companies like Google and OpenAI skew AI chatbots to reflect liberal biases, and are advocating for a more truthful, less politically correct approach to AI responses.

🤖 AI Chatbots Show Bias: Less Empathy for Certain Races

With over 150 million Americans living in areas with a shortage of mental health professionals, many are turning to the digital world for support, often posting on forums like Reddit to seek advice from strangers. Leveraging this trend, researchers from MIT, NYU, and UCLA used 12,513 posts from mental health subreddits to evaluate the quality and fairness of AI-driven mental health chatbots, like those powered by GPT-4. Remarkably, GPT-4 responses were found to be 48% better at encouraging positive behavioral changes than human responses, with licensed psychologists rating AI replies as more empathetic overall. However, the study also highlighted concerns, such as a Belgian man’s tragic death following a chatbot exchange and bias in GPT-4's empathy levels across different demographic groups.

Despite its promise, AI isn't without its flaws—GPT-4 was less empathetic toward Black (2-15% lower) and Asian posters (5-17% lower) compared to white or unspecified demographics. Researchers found that these biases could be reduced by explicitly instructing the AI to consider demographic context, which helped level the empathy playing field. While LLMs like GPT-4 outperform humans in some areas of mental health support, experts stress the need for careful oversight and thoughtful deployment. As AI becomes more integrated into clinical settings, the challenge remains to ensure it’s not only powerful but equitable and inclusive for all.

VC Fundraising Rounds

• Liquid AI Inc., an MIT spinoff specializing in generative AI models built on a novel architecture distinct from transformer-based systems, has secured $250 million in early-stage funding. (12/13/24)

• Zest AI, a company offering AI-driven credit scoring and underwriting solutions for financial institutions, has secured a $200 million growth investment. (12/13/24)

• Legion Technologies, a leader in AI-native workforce management innovation, has announced securing $50 million in financing. (12/12/24)

• Astrix Security, a trusted provider of enterprise solutions for securing non-human identities (NHIs), has raised $45 million in its Series B funding round. (12/10/24)

• WaveForms AI has secured $40 million in funding to advance its AI audio software, which enhances machine interactions by detecting emotional cues. (12/10/24)

Meme of the Day