• AI KATANA
  • Posts
  • New research addresses predicting and controlling bad actor AI activity in a year of global elections

New research addresses predicting and controlling bad actor AI activity in a year of global elections

Also: Google cancels contract with an AI data firm that’s helped train Bard

Author

AI KATANA
January 23, 2024

Welcome!

In today's edition of our newsletter, we delve into a spectrum of AI-related developments and their implications across different sectors. A study by George Washington University signals a potential rise in AI-driven misinformation during global elections, emphasizing the urgency for countermeasures. Meanwhile, the MIT Jameel Clinic's conference spotlights the critical need for regulating AI in healthcare, discussing FDA approval processes and ethical considerations. In corporate news, Google's termination of its contract with Appen, a key player in AI model training, marks a strategic shift in its business alliances. For businesses looking to harness AI, an MIT Sloan article offers valuable insights into identifying and prioritizing AI opportunities, aligning them with business objectives. On the tech front, Google Chrome's latest update introduces AI features aimed at enhancing user productivity and creativity, albeit with mixed reviews. Lastly, in venture capital news, Bluesheets, an AI automation software startup, secures US$6.5 million in Series A funding, poised for expansion in various sectors and geographical regions.

Sliced:

  • 🏛️ New research addresses predicting and controlling bad actor AI activity in a year of global elections

  • 🧑‍⚕️ What to do about AI in health?

  • ❌ Google cancels contract with an AI data firm that’s helped train Bard

  • 🤖 How businesses can find and prioritize AI opportunities

The research by George Washington University predicts an increase in AI-driven bad actor activities, particularly in manipulating information online during the election season. This study, the first quantitative analysis of its kind, highlights how even basic AI systems like GPT can be used for misinformation campaigns on social media platforms. It underscores the need for social media companies to deploy effective strategies to contain disinformation. The research provides insights into the what, where, when, and how of AI misuse by bad actors globally, emphasizing the importance of controlling such activities.

The MIT Jameel Clinic hosted a conference to discuss the regulation of AI in health, involving faculty, industry experts, and regulators. Key topics included the FDA's approval process, the need for a better understanding of AI mechanisms in healthcare, and concerns over AI's ethical use in clinical settings. The event highlighted challenges such as data availability, educating stakeholders, and the impact of AI on healthcare costs and labor shortages. The conference's goal was to inform regulators about AI advancements and explore new regulatory frameworks, emphasizing AI's growing role in medicine and healthcare.

❌ Google cancels contract with an AI data firm that’s helped train Bard

Google has terminated its contract with Appen, an Australian data company that played a crucial role in training its large language model AI tools, including those used in Bard Search. This decision is part of Google's broader strategy to reevaluate and optimize its supplier partnerships across Alphabet. Appen's collaboration with Google significantly contributed to its revenue, highlighting the importance of such partnerships in the AI industry. This move reflects the dynamic nature of corporate relationships in the rapidly evolving field of AI technology.

The MIT Sloan article discusses strategies for businesses to identify and prioritize AI opportunities. It emphasizes the importance of aligning AI initiatives with business goals and encourages companies to focus on areas where AI can provide significant value. The article advises businesses to evaluate the feasibility and potential impact of AI projects, considering factors such as data availability, technical capabilities, and alignment with strategic objectives. It suggests starting with pilot projects to test and learn before scaling AI solutions across the organization.

🛠️ AI tools updates

Google Chrome has introduced new AI features in its latest release to enhance productivity and creativity. These include a Tab Organizer for managing tabs, a feature for creating custom themes using AI, and a tool called Help Me Write to assist with text drafting on the web. These experimental features, part of Google's ongoing AI integration, aim to make browsing more efficient and personalized. However, there are mixed reviews from users and experts, with concerns about privacy, security, and accuracy.

💵 Venture Capital updates

Bluesheets, an AI automation software startup, raised US$6.5 million in Series A funding. The Singapore-based company, co-founded by Clare Leighton and Christian Schneider, uses proprietary AI software to digitize and automate bookkeeping for businesses in various industries. The funding, led by Illuminate Financial with participation from Insignia Ventures Partners, Antler Elevate, and 1982 Ventures, will aid in expanding into financial services, insurance, supply chain, and manufacturing. The capital will also support senior management hiring and grow Bluesheets' reach in Asia-Pacific and the U.S.

🫡 Meme of the day

⭐️ Generative AI image of the day