AI in Marketing

ChatGPT's mobile app experienced a significant increase in uninstalls, rising by 295% on February 28, following OpenAI's partnership with the U.S. Department of Defense, now called the Department of War under the Trump administration. This surge contrasts with the app's usual 9% daily uninstall rate, as reported by Sensor Tower. Concurrently, Anthropic's AI app, Claude, saw a 37% increase in downloads on February 27 and a 51% rise on February 28, after announcing its decision not to collaborate with the defense department due to concerns about AI's potential use in surveillance and autonomous weaponry. This decision appears to have resonated with consumers, as evidenced by Claude's ascent to the top of the U.S. App Store rankings on February 28. ChatGPT's download growth was adversely affected, with a 13% drop in U.S. downloads on February 28 and a further 5% decline on February 29. Additionally, ChatGPT received a 775% increase in 1-star reviews on February 28, with a subsequent 100% rise on February 29, while 5-star reviews fell by 50%. Appfigures confirmed Claude's download surge, noting it surpassed ChatGPT's for the first time. Similarweb reported Claude's U.S. downloads were 20 times higher than in January, though it suggested other factors might contribute to this growth beyond the political context.

Next.

Anthropic has introduced a new feature on claude.com/import-memory, allowing users to transfer memories from AI assistants like ChatGPT, Gemini, and others into Claude without losing data or starting anew. This process involves a simple two-step method: users paste a prompt into their current AI to extract stored memories, then paste the result into Claude's memory settings. This feature is available on all paid plans, ensuring that memories are encrypted, not used for model training, and can be exported anytime, avoiding vendor lock-in. Google's similar feature for Gemini only transfers chat logs, not memories, which means Anthropic's approach is more privacy-focused. The ability to import memories addresses the challenge of switching costs in the AI assistant market, where accumulated context, such as user preferences and work details, is valuable. Anthropic's move aims to attract users from competitors like ChatGPT by offering a seamless transition without the need to re-teach a new assistant. This development highlights the growing importance of memory portability as a competitive factor, with Anthropic positioning Claude as a primary AI assistant choice by reducing switching barriers and emphasizing user privacy.

On that note.

OpenAI's recent Pentagon contract has sparked criticism over AI safeguards and surveillance risks. CEO Sam Altman admitted the deal was rushed, following the collapse of Anthropic's negotiations with the Pentagon. Critics question why OpenAI secured the contract while Anthropic, which opposes using AI for autonomous weapons or mass surveillance, did not. OpenAI insists its models won't be used for such purposes, outlining safeguards like cloud-based APIs and oversight by cleared personnel. However, concerns persist, with critics citing potential loopholes in Executive Order 12333 that could allow surveillance involving Americans. OpenAI's Katrina Mulligan countered that system architecture, not just policy, prevents misuse. Altman defended the deal as a strategic move to ease tensions between AI firms and the Pentagon, despite backlash and competition from Anthropic's Claude.

Meanwhile.

NVIDIA and Coherent have announced a strategic partnership aimed at advancing optics technology to enhance next-generation data center architecture. This collaboration seeks to address the increasing demand for data processing and storage capabilities driven by the rise of artificial intelligence and machine learning applications. By integrating Coherent's expertise in photonics with NVIDIA's leadership in accelerated computing, the partnership aims to develop innovative solutions that improve data center efficiency and performance. The joint effort will focus on creating scalable optics technologies that can support the growing needs of modern data centers, ensuring they can handle the vast amounts of data generated by emerging technologies. This partnership represents a significant step towards optimizing data center infrastructure to meet future technological demands.

Shifting gears.

TechCrunch reports that the customer service industry is undergoing significant changes due to AI advancements. Startups like Decagon, Parloa, and Sierra are attracting substantial venture capital investments, while 14.ai, a Y Combinator-backed company, is pioneering an AI-native approach to customer support. Founded by Marie Schneegans and Michael Fester, 14.ai has secured $3 million in seed funding and aims to replace traditional customer support teams with AI-driven solutions. The company integrates its services quickly, managing support tickets across various platforms, and has demonstrated efficiency by rapidly clearing backlogs for clients. With a small team of six, 14.ai plans to expand its workforce with more AI engineers. The startup not only focuses on customer support but also aims to enhance sales and revenue growth by automating tasks. By reducing reliance on human labor and traditional systems, 14.ai positions itself as both a customer service and revenue growth engine, aligning with Y Combinator's vision for AI-powered agencies.

In other news.

Bruno Szarf, Global VP of Management & Performance at Stefanini, outlines strategies for responsible AI adoption, emphasizing the importance of culture over code. Initially, the introduction of AI tools like ChatGPT and Copilot raised concerns about data security and job redundancy. Stefanini addressed these fears by promoting transparency and demonstrating AI's role in enhancing, not replacing, human work. For example, AI tools were used to streamline resume screening, allowing recruiters to focus on candidate interactions. The company shifted from a defensive stance to encouraging guided experimentation with AI, ensuring data security through private instances and maintaining human accountability for AI outputs. Training was tailored to specific roles, moving beyond generic sessions to contextual learning. Peer influence was harnessed through initiatives like "AI Week," where employees shared practical AI applications. Leadership involvement was crucial, with executives actively participating in AI adoption, fostering a culture of curiosity and safety in the evolving AI landscape.

After that.

Iran is experiencing a near-total internet blackout amid a U.S.-Israeli cyber and military campaign, escalating tensions into a digital battlefield. Pro-Iran hacking groups have threatened Western and Gulf infrastructure, with 60 hacktivist groups reportedly active. The pro-Iran group Handala claimed a ransomware attack on Israel Opportunity Energy, signaling the start of massive cyberattacks. While no ransom details were provided, the breach narrative suggests data exposure. Qatar reported drone attacks from Iran on its energy facilities, causing damage but no casualties. The conflict has seen a surge in state-sponsored cyber activity, with Iran targeting U.S. energy and telecommunications infrastructure. The Islamic Revolutionary Guard Corps launched missile and drone attacks on U.S. military installations in the Persian Gulf, resulting in casualties and damage to assets. The U.S. and Israel have responded with air and sea strikes, targeting Iranian military infrastructure. Iran's "Mosaic Defense" doctrine enables decentralized, asymmetric warfare, with attacks on military bases, oil refineries, and data centers. The Strait of Hormuz has been declared closed by Iran, threatening global energy supplies. Internet watchdog NetBlocks reported Iran's connectivity at 1% of normal levels, with Cloudflare confirming a complete disconnection. Security researchers warn of increased cyber threats, with Iranian-aligned groups conducting reconnaissance and DDoS attacks. The U.S. Cybersecurity and Infrastructure Security Agency faces staffing challenges, raising concerns about its ability to respond to these threats.

Finally.

MCoBeauty, an Australian beauty brand, has launched an innovative campaign called "If You Know, You MCo" to engage its audience through a unique blend of entertainment and marketing. The campaign features a seven-episode microdrama series starring Tana Mongeau, where she navigates a beauty-themed escape room within MCoBeauty's offices. This concept is extended to real-life experiences at 30 Escapology locations across the U.S., where participants can engage in beauty lab escape rooms and receive free product giveaways. Meridith Rojas, CMO at VidaCorp, MCoBeauty's parent company, emphasizes the importance of creating an immersive brand world that goes beyond traditional marketing strategies. The campaign aims to build a tangible connection with audiences, moving away from passive consumption to active participation. MCoBeauty, known for its affordable dupe products, has been expanding its presence in the U.S. market with various initiatives, including appearances on reality TV and pop-up shops. The brand's strategy focuses on offering budget-friendly alternatives to prestige beauty products, embracing its identity as a dupe brand while adhering to legal guidelines to differentiate itself in the growing "masstige" beauty market.