AI in Marketing

PlusAI has launched SuperDrive 6.0, an advanced SAE Level 4 autonomous driving system designed for commercial freight operations. This release coincides with PlusAI's plans to go public following its acquisition by Churchill Capital Corp IX. The new system enhances capabilities such as night driving and construction zone handling, essential for 24/7 freight operations. CEO David Liu emphasized the significant advancements in predictive accuracy and AI development, which could expedite the deployment of autonomous trucks. SuperDrive 6.0 was developed using over 7 million miles of real-world driving data from the U.S., Europe, and Asia, bringing PlusAI closer to deploying scalable, driverless commercial vehicles by 2027. Currently, trucks equipped with SuperDrive 6.0 are already operational in Texas. The system's construction zone handling is active, with night driving expected soon. SuperDrive 6.0 also offers cost and deployment efficiency improvements, achieving a tenfold increase in AI training speed and a threefold reduction in data labeling costs through autolabeling, imitation learning, and reinforcement learning. This efficiency allows for quicker feature deployment and route expansion, enhancing scalability and revenue potential. The system features a distributed compute architecture using NVIDIA DRIVE Orin and Thor, ensuring performance under challenging conditions. A new Transformer-based "Reflex" layer in the PlusAI AV 2.0 architecture improves motion forecasting, doubling the accuracy in predicting the trajectories of dynamic actors, thus enhancing safety and decision-making in real-time operations.

Next.

OpenAI and Google DeepMind employees have filed a statement supporting Anthropic's lawsuit against the U.S. Defense Department, which labeled the AI firm a supply-chain risk. The Pentagon's designation followed Anthropic's refusal to allow its technology for mass surveillance or autonomous weaponry. The employees argue that the government's action is an improper use of power, potentially harming U.S. competitiveness in AI. They suggest the Pentagon could have canceled its contract with Anthropic instead of imposing the label. The Defense Department quickly signed a deal with OpenAI after the designation, prompting protests from OpenAI staff. The filing emphasizes the need for strong guardrails in AI use, given the absence of public laws governing the technology.

Meanwhile.

The Trump administration has unveiled a new cyber strategy emphasizing offensive measures and deterrence to address cyber threats. This seven-page document outlines a vision of American cyber power that combines domestic deregulation with aggressive actions against foreign adversaries. The strategy is not an implementation guide but a statement of posture and priorities, marking a departure from previous administrations' more prescriptive approaches. An executive order accompanies the strategy, aiming to disrupt transnational criminal organizations and cybercriminals targeting the US. A new operational unit within the National Coordination Center will coordinate federal efforts to detect, disrupt, dismantle, and deter foreign cyber threats. The strategy is organized around six policy pillars: detecting and disrupting adversaries, reducing cybersecurity regulations, modernizing federal networks, hardening critical infrastructure, ensuring leadership in AI and emerging technologies, and developing the cybersecurity workforce. The strategy's explicit focus on preemption sets it apart from earlier approaches that emphasized resilience and coordination. It aims to act against adversaries before they achieve their objectives, reflecting a more direct approach to cyber conflict. However, the strategy's vagueness and lack of specific implementation details pose challenges. It reads more like a high-level messaging document, lacking the specificity needed for decision-making. The strategy's brevity contrasts with the Biden administration's 2023 National Cybersecurity Strategy, which was more detailed and prescriptive. The Trump strategy rejects compliance-driven cybersecurity and frames AI as both a strategic asset and an attack surface. It includes aggressive deterrence language, emphasizing disrupting and imposing consequences on adversaries, with responses not confined to cyberspace.

On that note.

I joined Daniel Luquez on Entrepreneurship 101 today. The point I made most often: AI doesn't replace expertise. It amplifies whatever you already bring.

Anyone can ask ChatGPT or Claude to write a blog post. The output will be competent and forgettable. But if you know how to structure a narrative, inject context, direct tone, you get something worth reading.

The same applies to lead generation. At Nytro Marketing, we use AI to aggregate buying signals, hiring patterns, GitHub activity, press mentions, and surface the accounts most likely to convert. What used to take three hours now takes twenty minutes.

The tool changed. The judgment didn't.

With that said.

Anthropic has launched an AI tool called Code Review to address the challenges posed by AI-generated code, which has increased development speed but also introduced bugs and security risks. Code Review, integrated with GitHub, automatically analyzes pull requests, focusing on logical errors rather than style. It uses multiple agents to examine code from different perspectives, providing step-by-step explanations and prioritizing issues by severity. The tool is initially available to Claude for Teams and Enterprise customers and is designed to help large enterprises like Uber and Salesforce manage the increased volume of code generated by AI tools. Code Review offers a light security analysis, with pricing based on code complexity, estimated at $15 to $25 per review. Anthropic's enterprise business has grown significantly, with Claude Code's revenue surpassing $2.5 billion.

Moving on.

AI search has significantly altered how content is discovered, clicked, and converted, shifting from a visibility issue to a focus on measurement and budget. Many marketing teams are still adapting to these changes. SEJ Live, a free virtual event on March 11, aims to equip marketers with a Q2 plan through three sessions. The first session, featuring Matt G. Southern, Loren Baker, and Shelley Walsh, will analyze Q1's AI search changes, covering news, business impact, and content strategy. The second session, led by Emily Popson from CallRail, will address new KPIs for AI search, as traditional metrics fall short. Nikhil Lai from Forrester will conclude with insights on how answer engines are reshaping marketing strategies. The event includes live Q&A and an AMA for additional inquiries.

In other news.

Ellie Smith from Catalyst emphasizes the importance of leadership in successfully integrating AI into organizations. Despite significant investment in AI, many companies fail to focus on the people and systems needed to support technological change. Research by Catalyst and Coqual reveals that a third of employees fear being replaced by automation, leading to anxiety and mistrust. Organizations often prioritize AI investment over talent development, with fewer than one in three leaders planning for AI's long-term impact on people. The concept of a "convergent leader" is introduced, characterized by a balance of AI understanding, inclusive behavior, and sound judgment. Such leaders foster environments where AI is seen as augmenting skills rather than replacing them, leading to increased productivity, innovation, and retention. HR plays a crucial role in developing convergent leadership by redefining leadership criteria, supporting learning, and identifying naturally convergent talent. Effective leadership in an AI-driven world requires a blend of technology, trust, and human-centered decision-making.

Elsewhere.

OpenAI plans to acquire AI security startup Promptfoo to enhance its Frontier platform with agentic security testing and evaluation capabilities. Promptfoo's technology, used by 25% of Fortune 500 companies, helps identify and remediate vulnerabilities in AI systems. The integration aims to allow enterprises using Frontier to address risks earlier in development and meet governance, risk, and compliance standards. Srinivas Narayanan of OpenAI highlighted Promptfoo's expertise in securing AI systems at scale. Promptfoo will remain open source, continuing to support diverse AI models. Co-Founder Ian Webster emphasized the importance of securing AI agents as they connect to real systems. OpenAI launched Frontier in February, partnering with consulting firms and Amazon Web Services as its exclusive cloud provider.

Finally.

Partiful aims to become the default term for event invites, similar to how "Uber" is used for rides. The platform gained cultural recognition when mentioned in HBO's "The Pitt," surprising its founders. Partiful focuses on organic marketing, leveraging cultural moments rather than chasing viral trends. Its marketing team, though small, actively monitors trends and engages with users through features like secret crush identification. The app's user base primarily consists of Gen Z and millennials. Partiful has experimented with paid marketing and collaborations, such as with Lionsgate, but remains ad-free for now. The company emphasizes quality content, curating public events and publishing a newsletter to engage with party culture. Its strategy is to amplify existing cultural moments.