🔎 What’s good today:

• 💻A personal Data Analyst in your pocket

• 🦾 McKinsey: Human touch still matters in the media

• 🙈 We might have multiple personalities assistants in the future.

• 👀ChatGPT Traffic Drops

🔥A personal Data Analyst in your pocket

You've been patient, you've anticipated, and now it's here! OpenAI's new ChatGPT feature is ready to play its part in your life. Code Interpreter, as they christened it, may lack a bit of creative sparkle in its naming, but it carries an air of Sheldon Cooper-style precision that tells it all. I can't help but admire the straightforward simplicity of it.

Picture this, the Code Interpreter does exactly what it says on the tin. It serves as your personal data cruncher, your pocket-sized analyst. It can execute your code, decipher intricate charts, and solve the most convoluted math problems right there in your ChatGPT interface. If you're a user of ChatGPT Plus, things are now faster and sleeker. I give credit where it's due, this feature is a leap forward. But, let's be honest, is this an applause-worthy act or merely a polite golf clap moment?

Some of you may not share my enthusiasm, and that's fair game. Code Interpreter is something of a pacifier, a sweet morsel offered to soothe us after they retracted our beloved Bing Browsing. There's a certain disillusionment setting in as people begin to understand that these advanced features aren't trickling down to the free users as we'd hoped. So, yes, I get it if the spark isn't quite catching for you. However, let's keep an open mind and see where this new pathway leads us.

🤓McKinsey US Media Day review

Six months have elapsed since the birth of ChatGPT, a technology that propelled generative AI onto center stage and into the grasp of the public. This unassuming brainchild is poised to cause a seismic shift in the way we do business, and it's rapidly proving itself as a game-changer across all industries. And yet, the ramifications of its deployment remain enigmatic, sparking more questions than answers amongst leaders: How will this technology influence my business? What risks are on the horizon, and how do we tackle them? How does this reshape our perception of talent?

Such topics took center stage at a lively discussion hosted during McKinsey's US media day, fittingly titled the Halftime Report: A Mid-Year Update on the CEO Agenda. In the midst of business journalists, leaders engaged in a spirited exploration of the three primary themes reverberating throughout the industry: generative AI, the future of work, and the paramount role of productivity in propelling economic growth. Ben Ellencweig, the Senior Partner leading alliances and acquisitions at QuantumBlack, noted that businesses are rapidly learning about generative AI, being compelled to adapt to this groundbreaking technology at breakneck speed.

The conversation, steered by Axios’s global-technology correspondent Ryan Heath, revolved around what Ben termed "the four Cs": coding, customer engagement, creative content generation, and content synthesis. He emphasized that generative AI is emerging as the "virtual knowledge worker," capable of connecting disparate data points, summarizing and synthesizing insights in mere seconds, and ultimately liberating us to focus on more high-value tasks. Meanwhile, Jacky Wright, McKinsey's Chief Technology and Platform Officer, posited that the automation of mundane or repetitive tasks will catalyze a surge in more cognitively focused pursuits.

Wrapping up the media day, the conversation pivoted towards productivity, the bedrock of US economic power and prosperity since the Second World War. Despite having seen a dip in growth in recent years, the belief persists that improvements in labor productivity remain critical. Asutosh Padhi, Senior Partner and North America managing partner, put it succinctly: "Productivity is not an answer to growth and economic prosperity… it is the only answer.” As we continue to dissect and understand generative AI's potential, it is crucial to keep in mind that the ultimate power rests with us. As Alex Singla, QuantumBlack Global Leader, reminded us: "For most generative AI insights, a human must interpret them to have impact. The notion of a human in the loop is critical.”

👨‍🚀AI Assistants with Multiple Personalities

Chatbots are advancing from mere draft writers to empathic personal guides. Consider Pi, a chatbot developed by Silicon Valley startup, Inflection, that recently helped a man relinquish his smoking habit. This highly personal approach to AI has given Inflection quite a leg up, having raised an impressive $1.3 billion to develop a “personal AI for everyone”. Such advancements have sparked a compelling debate among industry bigwigs – do we lean towards anthropomorphizing artificial intelligence like OpenAI’s ChatGPT, or do we strip it back, creating a more neutral, functional tool akin to a smartphone operating system?

In this new world where the line between humans and their digital counterparts blurs, it's likely we'll rely on both styles of AI to optimize our productivity and guide us through our personal lives. As Mustafa Suleyman, Inflection co-founder, envisions, Pi could act as a personal chief of staff, a role that aligns closely with your interests, accompanies you on your journey, and offers constant feedback and advice. Even as AI ventures deeper into our personal lives, there's a crucial balance to strike: while Pi is patient and supportive, it's also designed to be clear that it's not human and doesn't possess feelings.

The rising interest in companion-style AI may also reflect our societal landscape, where the remnants of the Covid-19 pandemic continue to reverberate, leaving many more isolated than ever before. AI builders like Suleyman and Reid Hoffman predict that the trend of engaging with AI on a personal level is just the beginning. However, as San Francisco startup Adept explores, the human-AI interaction can be functional and more hands-on, a natural language interface for our computers. Companies like OpenAI are seemingly straddling both styles, creating an entity like ChatGPT to converse with, and a more work-oriented system embedded in our infrastructure. Regardless of the direction, it seems inevitable that our dialogues with computers will only grow in the future.

👀 ChatGPT Traffic Drops Nearly 10% In June

Preliminary data estimates from Similarweb reveal that worldwide web traffic to the ChatGPT website, chat.openai.com, declined by 9.7% in June compared to May, with a 10.3% decline in the U.S. The number of unique visitors to the site also fell by 5.7%. Despite the decrease, ChatGPT still attracted more visitors worldwide than Bing.com and Character.AI.

Character.AI, a chatbot site that takes on the personalities of celebrities and fictional characters, experienced a 32% decline in worldwide visits compared to the previous month, although traffic remained higher than in June 2022 when the company was newly established. Both ChatGPT and Character AI have encountered challenges such as content access loopholes and copyright lawsuits related to training their models. Additionally, the energy-intensive nature of training powerful AI models like ChatGPT raises concerns about environmental impact, with a study suggesting that training GPT-3 consumed 185,000 gallons of water, equivalent to the water needed to fill a nuclear reactor's cooling tower.

That's that for today! See you next Tuesday!
Tony