Coactive AI at AWS re:Invent: Conference Recap
The 2023 edition of AWS re:Invent recently took place in Las Vegas, attracting over 50,000 attendees. re:Invent is one of the largest cloud-computing and learning conferences in the world, hosted by Amazon Web Services (AWS). At the conference, our cofounders had the opportunity to meet with Howard Wright, VP, Global Head of Startups at AWS, during the conference. Our relationship with AWS extends beyond platform support; we are an advanced tier partner on the AWS Partner Network (APN) and actively engaged in their marketplace. Our first attendance at the event brought key insights from notable keynotes, product announcements, and conversations with customers and peers.
Coactive AI leads in revolutionizing multi-modal asset management, enhancing the search and analysis of unstructured audio, video, and image content. Our objective is to simplify how insights are gained from multi-modal content, a key challenge for enterprises managing expanding content libraries.
Seth Mazow, on our business development team, succinctly summarized the conference's theme: "The time for AI is now," highlighting AI's readiness to transform the handling of underutilized visual content in organizations. He pointed out the limitations of traditional tagging systems and the emerging solutions AI offers. Echoing this, CEO Cody Coleman predicted that 2024 will be pivotal for AI's shift from prototype to production, underscoring the timeliness of our participation in the conference. This blog post delves into Coactive AI’s experiences at the event and our insights on AI's role in enhancing operations for organizations dealing with user-generated and media based content.
Product Highlights at re:Invent
Amazon Bedrock's announcement at AWS re:Invent 2023 highlights its commitment to enhancing AI model compliance and practical application. Bedrock now features guardrails for model compliance, customization capabilities, and Retrieval-Augmented Generation (RAG) for better integrating internal enterprise data. Its Continued Pre-Training and newly introduced agents that help the end-user complete actions based on their organizations data and inputs.
Echoing the importance of building upon foundational models, William A. Gaviria, one of the co-founders of Coactive AI, said "There's a lot of people that are not looking for foundation models. They're looking for the solutions for what the foundational models do... they should have something called “Amazon topsoil” to answer the fundamental question of , what you can actually do on top of the Bedrock" This observation underlines the necessity of developing accessible, application-focused tools on top of foundational technologies like Bedrock, addressing the needs of the many potential users who do not have deep technical expertise.
Further, the introduction of Amazon Q, a generative AI assistant, closely aligns with our vision at Coactive AI. Integrating this technology enhances our ability to make your businesses’ unstructured data usable, a core aspect of our mission. Furthermore, the AWS Graviton4 and Trainium2 chips represent a significant leap in running AI and data-intensive workloads. Will’s insight, "Helping humans and machines work together... our view is that AI is just going to unlock all of the full creative potential of humans,” reiterates our excitement for these developments. For a full outline of the major product releases check out this recap from Amazon.
Challenges and Opportunities
The conference spotlighted the challenges in scaling AI. Cody pointed out, "It's uncertain which AI use cases will thrive in production," steering our strategic focus towards scalability, security, and efficiency. Here are some of the key takeaways from the conference:
Economic transformation through AI
The current transformation in AI technology represents a significant economic shift, similar to the impact of the invention of the internet. Will captures this by stating, "The transformation is not only for business technology, but also for the economics" This highlights the profound changes AI is bringing to the ways in which industries operate and innovate.
As AI reshapes the business landscape, organizations face the challenge of integrating these new technologies effectively. Will notes, "The change is going to touch every industry the same way the internet did. Organizations are going to have to figure out how to implement it." This underlines the universal impact of AI across various sectors, signaling a period of learning and adaptation.
Leading this wave of change are startups, and particularly in the areas of image and video technology stands Coactive AI. Will emphasizes the company’s leadership role: "We are going to be leading the way in defining what the new AI-focused tech stack looks like when it comes to utilizing image and video." This statement positions startups like Coactive at the forefront of leveraging AI - not just following trends but actively defining them.
Increased budgets and focus on AI within enterprises
The transition to real-world AI solutions is gaining momentum, especially as enterprises gear up with dedicated resources and strategic mandates for AI implementation. The shift from prototyping to production in 2024 is underpinned by a newfound corporate readiness to embrace AI, as evidenced by increased budget allocations and board-level mandates. "Every board in the world knows about AI and is thinking about it... companies have mandates from boards to do stuff with AI and budgets to go along with it," Cody remarked. This change is largely driven by the realization of AI's potential following major developments like ChatGPT, which caught many by surprise in late 2022. As enterprises ready themselves to catch up with the consumer-level AI demand, the stage is set for a significant leap in AI application in the business world.
Challenges in implementing AI at scale
As companies transition from prototyping to deploying AI solutions in real-world scenarios, they face numerous challenges, particularly in ensuring security, scalability, and efficiency. The anticipation for AI's move from prototype to production is clear, yet there's an acknowledgment of the complexities involved: "It's unclear which ones are actually going to survive when we think about production loads because of security, scalability, and just the efficiency," Cody explained. This sentiment reflects the hurdles in scaling AI technologies to meet the demands of production environments. Enterprises are keenly aware of the potential pitfalls and are cautiously optimistic as they navigate these challenges to harness AI effectively.
Trend towards model-agnosticism in AI
The AI landscape is witnessing a proliferation of models, leading to a trend towards model agnosticism. This trend is captured by Cody’s observation that, "There's been an explosion of models... it's unclear which one's going to rule them all." This reflects a growing understanding that no single AI model is universally superior, paving the way for more diverse and tailored AI solutions.
Conclusion
AWS re:Invent 2023 was more than just a conference for Coactive AI; it was an opportunity to share ideas and connect with the top innovators in cloud technology. As Seth aptly put it, “Attending these events is an investment, not just a cost.” We left re:Invent invigorated, ready to embrace the opportunities and challenges ahead.
Excitingly, one of the highlights for us was Cody’s interview with Sam Charrington, the host of the TWIML AI Podcast, a respected voice in the AI community. Cody delves into Coactive AI's use of modern data, systems, and machine learning techniques, including active learning and core set selection, to enhance our multimodal asset platform and visual search tools. You can listen to the full episode here.
As we’ve seen in this blog post, Coactive AI’s unique way of combining human intelligence and machine learning has created new standards in how models can be developed, managed, and monitored. This will not only change the way businesses operate but also provide better insights into the decision-making process that can be used to improve operations.
Stay connected with us by following us on social media. As part of this evolving AI landscape, we're excited for what the future holds.
