Decoding tomorrow's tech, today

Dear tech friends,

Welcome to the latest edition of?BCG X Files, our monthly tech-focused BCG X LinkedIn newsletter.

As we're approaching the one-year anniversary of ChatGPT, the day when generative AI took the world by storm, we can admit that it has undeniably changed the world as we know it. But is it really that new, or is it just another form of something we've already experienced before?

While we witness generative AI causing a profound disruption to industries and businesses and helping solve many complex problems facing the world today, there are still many misconceptions and myths that come along with this groundbreaking technology.

In our new video series, 'GenAI Explored,' together with Leonid Zhukov, Ph.D , Leader of BCG Global AI Institute, an R&D team within BCG focusing on emerging AI technologies, we delve into the depths of GenAI and shed light on the realities shaping this cutting-edge technology.?

The first two episodes are already live on LinkedIn, along with our full version, waiting for you on YouTube. Be sure to check it out and stay tuned for the upcoming episodes!

Want to get the BCG X Files in your inbox every month? Just hit the subscribe button. Feel free to share your thoughts and ideas with us anytime in the comment section below.?

Enjoy!?

-Sylvain Duranton, Global Leader of BCG X

How Generative AI Can Boost Highly Skilled Workers’ Productivity | MIT Sloan Management Review - 5 min read

A study by BCG Henderson Institute found that generative AI can boost productivity by 40% within its capabilities but decreases it by 19% for tasks beyond. To optimize, workers should apply judgment, not blindly follow AI. Orgs should train on AI limits, promote role experimentation, and foster a culture of accountability with or without AI.

Generative AI Throwdown: Open Source Vs. Proprietary Models | Forbes - 3 min read

A new wave of open-source models is disrupting the field of GenAI. Some argue that open sourcing offers essential advantages, such as expanding accessibility, driving innovation, and advocating for transparency. However, generative AI demands resources in terms of data and energy, both of which are becoming progressively costly and inaccessible for many open-source players.

Putting Medtech People and Processes to Work with GenAI | BCG - 9 min read

GenAI will drive innovation in healthcare using sensitive patient data. It can improve efficiency and tackle challenges like rising costs and worker shortages. About 18% of the global workforce (300 million people) will be affected by GenAI in the next decade, primarily automating repetitive tasks.

Strange Ways AI Disrupts Business Models | The Implications Substack - 12 min read

Scott Belsky, Adobe's Chief Strategy Officer, delves into AI's transformative effects on business models and industries, highlighting the shift from data analysis to real-time optimization. This change may require adaptations in industries and business models, impacting time-based models, decision-making in purchases, entertainment creation, and traditional creative networks.

How AI Could Supercharge Battery Research | MIT Technology Review - 5 min read

At the MIT ClimateTech event, Venkat Viswanathan, co-founder of the startup Aionics, explained that AI has great promise to speed up battery research by enabling researchers to screen far more potential materials combinations than possible through traditional trial and error experimentation alone, with the goal of developing better electrolyte materials for EVs and other technologies in a faster timeline.

Why a 'Hologram Revolution' Could Be on the Way | BBC - 4 min read

The Hologram Zoo in Australia uses lasers to project lifelike animal holograms. Researchers are working to miniaturize hologram projection and apply it to fields like medicine and augmented reality. Holograms could soon be used more widely in applications like fast food restaurants and retail stores.

New Data from Cruise on Safety Results Regarding Autonomous Vehicles on the Road | Cruise Blog - 7 min read

The Cruise team presents research from UMTRI and VTTI that analyzed over 5.6 million miles of human ridehail driving data to establish a baseline crash rate for human drivers in dense urban environments. This baseline human ridehail crash rate of 1 crash per 15,414 miles was then used by Cruise to benchmark the safety of their driverless fleet. Cruise also performed additional analysis to identify subsets of crashes involving meaningful risk of injury and primary responsibility to provide more granular safety metrics.

Is Your Organization Investing Enough in Responsible AI? ‘Probably Not,’ Says Our Data | MIT SMR - 8 min read

BCG and MIT Sloan Management Review have once again gathered global AI experts to explore RAI implementation. This year, we asked academic and practitioner panelists whether companies are investing enough in RAI as they become more aware of AI's risks.

China Has a New Plan for Judging the Safety of Generative AI—and it’s Packed with Details | MIT Technology Review - 4 min

China has proposed new detailed rules for how companies should evaluate the safety and censorship compliance of generative AI models, including specific metrics for assessing training data quality and requirements for large content moderation teams. These proposed standards provide clarity on how China will regulate generative AI according to its laws, but may also influence global AI development.

The US Has Failed to Pass AI Regulation. New York City Is Stepping Up | WIRED - 5 min read

The US government has failed to pass meaningful regulation of artificial intelligence. In response, New York City introduced an AI Action Plan with 40 policy initiatives to protect residents from harm by AI. A New York City council member also introduced legislation to create an Office of Algorithmic Data Integrity to oversee and assess AI systems used by the city government.

VC Firms Working with D.C. to "Self-Regulate" AI Startup Investing | Axios - 2 min read

Dan Primack covers an effort by major venture capital firms to develop self-regulatory guidelines for RAI. Dozens of firms are working with the US Commerce Department to create standards around issues like transparency, safety testing, and ongoing improvement. The goal is to encourage ethical "AI cultures" in startups from the beginning in order to head off stricter government regulation later.

Google Wants to Be at the Center of all Your Climate Change Decisions | The Verge - 7 min read

Google announced several new tools and updates aimed at helping individuals, organizations, and governments make more sustainable choices to mitigate climate change, including enhanced search results for EVs, home appliances, and routes. They are also expanding existing tools to provide more accurate wildfire, flood, and heat warnings. The overall goal is for Google to be central to people's climate decisions by providing trusted information and data to drive lower emissions choices.

Artificial Intelligence is Key to Better Climate Models, Say Researchers | CalTech - 5 min read

Researchers at CalTech found AI could help improve climate models by generating data to inform parameterizations in coarser models. They can use machine learning algorithms trained on climate data to represent small-scale processes that can't be directly simulated. They believe this approach can create robust climate models without requiring exponentially more computing power for higher-resolution global simulations.

The Food Industry is Using AI to Create Healthier, Tastier Snacks and Meals. It Could Super-Charge Nutrition in America | Fortune - 4 min read

Food companies are using artificial intelligence to make foods healthier and more nutritious. A company called January AI uses data from continuous glucose monitors and activity trackers to create "digital twins" of people and predict how different foods will affect their blood sugar levels. The startup, already working with partners like Nestle, helps food companies reformulate products with this data to be healthier for people with diabetes, pre-diabetes, and the general population.

We hope you enjoyed this edition of BCG X Files. Don't miss out on the next one - subscribe today!