Episode Categories:
Resources:
Join the discussion on LinkedIn: Got something to say? Let us know on LinkedIn and network with other AI leaders
Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup
Connect with Jordan Wilson: LinkedIn Profile
Start Here Series in our Inner Circle Community: Join for free access
Open Source AI in 2026: Cost, Capability, and Legal Trade-Offs for Strategic Adoption
AI strategy for businesses is no longer a binary choice between a handful of well-known proprietary vendors. The field of open source AI models has accelerated dramatically, forcing enterprises—especially those with significant AI investments—to revisit foundational decisions about model selection, cost structures, and legal exposure. Recent developments have collapsed the historical performance gap and created dynamic new opportunities (and risks) around how organizations deploy AI at scale.
This article draws from the detailed analysis in the latest "Everyday AI" episode, translating concrete developments, pricing changes, and legal realities into actionable insights for business leaders who must navigate this complex terrain.
Open Source AI Models: Closing the Performance Gap With Proprietary Platforms
Historically, enterprise AI adoption defaulted to closed, proprietary models (e.g., from OpenAI, Google, Anthropic), primarily due to large capability gaps versus open-source alternatives. As recently as two years ago, open source models lagged the "frontier" by a significant margin—ELO score gaps were as high as 250 points, making open solutions a non-starter for most business-critical applications.
Today, performance parity is nearly achieved. The best open source models now lag their frontier counterparts by as little as 10–30 ELO points—often indistinguishable for all but the most intensive use cases. On consumer hardware, organizations can run models at a level comparable to what was previously reserved for multi-million-dollar, cloud-only contracts just over a year ago. This rapid progress was propelled by both Western AI labs (notably, Google’s Gemma 4) and expanded by legally controversial practices such as model distillation from leading Chinese labs.
Key Value: Business decision makers now have credible open source models as alternatives for many routine and even advanced workflows, often at a fraction of prior costs. This introduces immediate and substantial savings potential, especially for high-volume, repetitive tasks.
AI Cost Structure: From Expensive APIs to Pennies-on-the-Dollar Alternatives
Enterprise AI budgets have historically accommodated significant SaaS and API spends—figures easily stretching into seven- or eight-digit annual line items for large organizations. The entrance of powerful open source models, some developed by distilling outputs from leading US models, has disrupted this value chain.
For instance, Chinese open source providers such as DeepSeek now offer API access at prices as low as 43¢ per million token inputs and 87¢ per million token outputs. This is over 25 times cheaper than the most advanced closed-source proprietary options. Major enterprises have already reported realistic scenarios for saving millions merely by switching non-sensitive workloads (e.g., summarization, classification, extraction, parsing) to these alternatives.
Key Value: Enterprises with heavy AI utilization (e.g., those previously spending $40,000 per year per advanced AI seat) can realize order-of-magnitude cost reductions by triaging workloads—shifting high-volume, low-risk tasks to open source models or low-cost APIs, while reserving expensive frontier models for specialized needs.
Local AI Agents: Always-On Automation at Zero Marginal Cost
The advance of open source models has enabled always-on, local AI agents running continuously on commodity hardware. Google’s Gemma 4 model, for example, is 20 times more efficient than previous class leaders, allowing organizations to run near-frontier performance models directly on consumer laptops (e.g., recent MacBook Pros) with no ongoing marginal cost after initial download.
This shift enables organizations to deploy 24/7 local agents for a range of applications, from content generation and summarization to more complex, agentic workflows, without incurring per-use API charges.
Key Value: This transition unlocks autonomous, uninterrupted AI assistance—reducing dependence on cloud connections and subscriptions, and affording both cost containment and heightened data privacy for internal operations.
Legal Risks: Loss of Indemnity and Regulatory Compliance
The move to open source is not without trade-offs. Proprietary vendors such as OpenAI, Google, Microsoft, and Anthropic typically offer enterprise customers legal protections, including indemnity on intellectual property and regulatory compliance for outputs used in commercial products or customer-facing solutions.
Open source models, distributed under licenses such as MIT or Apache 2.0, generally disclaim warranty and non-infringement, leaving organizations exposed to downstream risks. For regulated work or customer-facing outputs—such as in finance, healthcare, or other compliance-driven sectors—this missing legal shield can translate into unmeasured liability, potentially negating cost savings if an incident occurs.
Key Value: Cost-saving measures must be carefully balanced against the absence of legal protection in open source solutions—a factor that remains non-negotiable for regulated and high-liability enterprise applications.
AI Model Triage: Strategic Workflow Segmentation
The new best practice is not to pick a single AI platform but to segment workloads by risk, value, and compliance obligations. Production environments should be viewed through the lens of triage:
High-volume, low-stakes tasks: Summarization, internal research, content generation—candidates for open source models/local agents.
Medium-risk, private workloads: Internal processes with privacy needs (e.g., confidential document parsing) may benefit from bespoke self-hosted models.
High-value, customer-facing, or regulated tasks: Retain proprietary, indemnified models to maintain compliance and protection.
Key Value: Structured approach enables maximization of savings and operational efficiency without forgoing core legal/business safeguards.
Outlook: Proliferation of Specialized, Self-Improving Local Models
Emerging trends point toward the rapid proliferation of small, specialized local models capable of recursive self-improvement. This will enable organizations to further customize and optimize AI deployments for niche processes, maximizing both efficiency and competitive differentiation.
Key Value: Early awareness and a proactive approach to integrating and governing open source and local models will position organizations for greater cost agility, technical flexibility, and robust risk management in an AI-saturated market.
Summary:
Open source AI models now stand as credible, cost-efficient alternatives for many enterprise use cases, with performance parity closing rapidly due to both innovation and controversial distillation practices. The economics of AI have shifted, enabling automation and autonomy unimaginable just two years prior. However, prudent organizations must balance these advantages with legal realities—deploying a triaged, workload-by-workload AI strategy that preserves compliance and business integrity.
Topics Covered in This Episode:
- Open Source AI vs Closed Models Shift
- Chinese Model Distillation & Legal Impacts
- Enterprise AI Cost Triage Strategies
- Google Gemma 4 Local Model Capabilities
- Frontier Model Performance Gap Closing
- 24/7 Agentic AI Systems Overview
- API Pricing War: DeepSeek vs US Vendors
- Legal Protection Tradeoffs for Open Source AI
- AI Workflow Triage: Task-Specific Models
- Future Trends: Local and Specialized LLMs
Episode Transcript
Jordan Wilson [00:00:19]:
Meet Firefly AI assistant, now live in Adobe Firefly, the all in one creative AI studio. Just describe what you wanna create, and the assistant handles the rest, orchestrating multi step workflows across Photoshop, Premiere, Express, and more in one conversational interface. You direct the outcome, the assistant accelerates execution. A few weeks ago, the United States government said the quiet part out loud when it comes to open source models, at least from China. That's because in April, the White House sent out an official memo accusing China of using distillation to illegally copy American AI models to create cheaper domestic knockoffs. And that declaration is really nothing new if you've followed AI for years. However, the recent distillation trend has completely reshaped one important landscape of enterprise AI, the decision between using open source models versus proprietary closed models. And in 2026, at least, it can actually be a tough choice between saving potentially millions of dollars versus running up your legal liability.
Jordan Wilson [00:01:39]:
About two years ago, before Chinese distillation was commonplace, there was a sizable gap between frontier models and open source AI models or those models that you can essentially download or use for close to free. But now the gap is all but closed, which has thrust the open source versus closed source question into every enterprise boardroom in 2026. And although there's no one size fits all answer, we're gonna be tackling the toughest topics and the most important takeaways as we take a zoomed out view of open source models on today's show. That's why we're going over open source AI one 2001, why local models, cheap APIs, and AI agents change everything about making AI decisions in 2026 as part of our start here series. Alright. Welcome to everyday AI. Before we dig in, let's first zoom out and talk about the big picture here when it comes to open source AI. That's because, well, it's actually a legitimate thing now.
Jordan Wilson [00:02:45]:
Right? Two years ago, enterprise companies weren't saying, let's use an open source AI model in production. Today, it's actually happened. It's happening. That's because, you know, maybe the most powerful open source models are only about two to six months behind frontier models. But on even consumer hardware, you can be running essentially frontier level AI models from, like, just over twelve months ago. And the Chinese labs now distilling US models have kind of crashed the open API prices to pennies. Right? So, yeah, not everyone out there on, you know, consumer or prosumer hardware can run the most powerful open source models. Although, I do think Google has something to say about that.
Jordan Wilson [00:03:33]:
But the most powerful open source models run for a fraction of the actual cost if you can't afford to run them locally, which has completely shifted the paradigm when it comes to enterprises making decisions on, well, are we gonna use a model from one of the big three, OpenAI, Google, or Anthropic, or are we gonna use a Chinese open source model and pay for it that way? And, well, what this has also led to in 2026 is essentially twenty four seven local agents that can run, and also without costing a ton and now having to actively and almost aggressively, aggressively go through kind of an AI cost triage. But going full open source does strip away the legal protection that the closed models often include. So stick with me for twenty five ish minutes on today's start here series show, and here's what you're gonna learn. You're gonna know why the open versus closed source AI default just officially flipped. You're gonna know how Gemma four from Google puts year old frontier capability on your laptop. You're gonna understand the two payoffs already shaping and reshaping how individuals and enterprises run AI and the hidden legal trade off most executives miss when going fully open source. Let's get into it. My name is Jordan Wilson.
Jordan Wilson [00:04:58]:
Welcome to Everyday AI's start here series. This is the essential podcast series to learn the AI basics. And if you're an AI expert, this is your chance to freshen up and double down on your AI knowledge. Why do we start this start here series? Well, after 750 plus podcast, I never really had a good answer when someone was like, where do I start? What podcast do I start with? That's why we created the start here series. It's best, I think, if you listen in order. I think this is now volume 24 of the start here series. So maybe we'll wrap it up at twenty five. Maybe we'll wrap it up at thirty.
Jordan Wilson [00:05:33]:
I'm not sure. But the whole point of this is you can go to starthereseries.com. That's gonna give you free access to our exclusive inner circle community. And in the start here series space, we make it even easier for you. So you can actually go listen. We have a Spotify playlist ready for all of the different start here series shows as well as a breakdown on each individual episode all in one place. So make sure you go to starthereseries.com for exclusive access to that inside of our inner circle community. Alright.
Jordan Wilson [00:06:02]:
And if you miss our last start here series show, that was volume 23. We talked about headless software and why companies are building software for AI agents and not humans and, well, what that means. So today, in volume 24 of the start year series, we're going over open source AI one zero one. So here's the reality. Closed AI used to be the de facto. Right? And I I mean, honestly, there was really never even much of a discussion about open source AI in the enterprise maybe until, 2025, at least not serious enterprise companies. Now it's a it's a real conversation. Right? So it's no longer, you know, hey, we're just gonna choose whichever API works best for us.
Jordan Wilson [00:06:48]:
Right? Whether that's OpenAI, Anthropic, or Google. Now most companies are looking at some of the open source alternatives. Most of them coming from China. And the big kind of thing here is companies are starting to standardize around one frontier vendor, in 2025 before this happened, and then they called it an AI strategy. And the assumption originally was, well, that worked for three years until two very specific forces broke the standard paradigm when it came to open source models. First, the proprietary versus closed gap capabilities just completely changed. Alright. So we talk about arena on here a lot.
Jordan Wilson [00:07:35]:
For, previously had a different name. Now it's just arena. So you put in a prompt. You don't know the outputs, you know, what models they're from, and you vote for which one is better. Right? So all these different models get an Elo score. And to really zoom out for our nontechnical audience, because I know a lot of you in the start here series are not technical, I'd probably even say what an open source model even is. Right? So, the very simplified version is certain companies can release models open source under, like, an MIT, or Apache two point o license, and that gives people the ability to download these actual models and to run them locally on your machine. And that is, well, one of the big trade offs.
Jordan Wilson [00:08:15]:
Right? So you're not sending any private or potentially proprietary data, in the cloud at all. Everything runs locally on your machine. So number one, it's private. Number two, it's, well, free. Right? And then there are open source models that if you can't download them on your, you know, computer, because not everyone can. Some of them are much larger. You can still essentially run those in the cloud for a fraction of the cost of what it would, cost to run a proprietary model. So, essentially, open source models are ones that you can download, you can modify.
Jordan Wilson [00:08:46]:
In some instances, you can even build products on top of it. Alright. Anyways, right until late twenty twenty five, there was a monstrous gap in the arena scores. Right? So these Elo scores, when you put in the same prompt, you look at two outputs, everyone overwhelmingly always chose the best front, the best frontier closed source model. And and that really started to change. Right? So the gap between the Elo scores, well, it cut down by about 90%. So it went from about a 250 gap from the best, Frontier or closed source model. Well, now it's only about 30 points.
Jordan Wilson [00:09:25]:
Right? Give or take, depending on the day. Right? But it's even, you know, recently, couple months ago, it was, like, 15 points. Right? So at that point, you really have to be an AI expert to be able to, decipher the difference. I think 30 points, you know, most people could look at different outputs over time and, you know, a 30, you can kind of realize it if you're looking at the best, you know, open source model versus the best proprietary closed models, 30 points you can understand. But 10 to 15 points, it's kind of a coin flip even for people who are, you know, spending most of their days inside of large language models. But this collapse came from those two different forces working in parallel. So I have a little, little graphic here on the screen for our livestream audience. If you are listening on the podcast, FYI, you can always get the video version on our website at youreverydayai.com.
Jordan Wilson [00:10:19]:
But going from a 250 gap to essentially a 30 gap, this is huge. Right? Because, like I said, in 2023 to mid twenty twenty five, it was noticeable. Right? It was extremely when you looked at the outputs, you could say my business can use output a, but it cannot use output b. Right? And now we're at the point where the open source models, in terms of an ELO score, and I think that's a good metric to look at over time. Right? Because, you know, the frontier is always improving. But if you look at the ELO scores of the open source models now, so those that you can kind of, you know, if you have a beefy enough computer, you can download some of the best open source models on your, actual local machine. Right. Those scores are where we were at with proprietary models three to six months ago.
Jordan Wilson [00:11:16]:
Right? So think back to the very end of 2025, you know, and there's some models, like, I think at the time was probably, GPT five three, Gemini three Pro. And, at that at that time, I think we're at, like, Opus, maybe four, six, or maybe four, five. Right? Now you have open source models that you can run for free twenty four seven, run agentically that are at that same level. And that's why now this is a real enterprise boardroom problem, especially for large companies that have invested heavily, into AI. Right? So I'm not talking about companies that, you know, with a couple 100 employees. I'm talking about companies that were spending 7, maybe 8, maybe even more 7 to 8 figures on AI each year. Now all of a sudden they're saying, hey. In theory, if we switched, you know, part of our, you know, summarization tasks alone.
Jordan Wilson [00:12:12]:
Right? There's I've I've read a lot. I've I've talked with a lot of people that have done something similar. You know, if if if we just, you know, chunk off everything that we're using, you know, open source model or sorry. Closed source model just for summarizing text. Right? Some of those lower hanging fruit. People are saying, well, yeah, we could save $1.02, $34,000,000, and this is an actual reality that a lot of companies are, grappling with right now. So the force one was just the capability moving locally. Right? And this, I think we have to credit Google for pushing the edge of edge AI.
Jordan Wilson [00:12:55]:
That's because with their Gemma four model, completely shook up, the landscape of open source AI. Right? This thing was 20 times more efficient than other open source AI models at the time. So, essentially, let me describe it like this. Do you remember g b t four o? Right? One of the best models, you know, about fourteen, fifteen months ago. It was at the absolute frontier. Right? So now you can download Gemma four on a consumer laptop. And it has you know, roughly, you look at the scientific benchmarks and the Elo score. It's essentially about a GBT four o level model that you can run on your laptop.
Jordan Wilson [00:13:45]:
So what's the big deal? What's the big difference? Right? Fourteen, fifteen months ago, I mean, there's thousands of companies spending millions of dollars a year to get that type of technology, to get GPT four o level technology for their employees. Now doesn't take take anything, really. It takes a new ish piece of Apple. Right? I I just got a a new MacBook Pro. That thing can run Gemma four very easily. Right? It can run even better models than that. And and this really changes, I think, what is ultimately capable when you look at the open source versus closed source. Because, yes, I think most people look at normal usage, and they're comparing apples to apples.
Jordan Wilson [00:14:32]:
Right? Here's what our marketing team did fifteen months ago with a GPT four o level model. Oh, now they can do that on Gemma four. Well, yeah. You can do it, but now you can do it agentically. Because not only in the last year or so have the models obviously improved with now thinking models, reasoning models being the default. But now we have these agentic harnesses. You know, not just the ones that you can use, you know, inside of ChatTBT, Gemini, Claude Copilot, but, well, you have these local, autonomous AI systems as well, such as OpenClaw, such as, I always forget if it's Hermes or Hermes agents. Right? So now you can have, essentially, the level of AI from fourteen months ago running for free twenty four seven agentically.
Jordan Wilson [00:15:28]:
Even if you're just doing it, you you know, summarization, content creation, research, things like that. So when capabilities went local, right, Gemma Ford leading the way, but, obviously, all the Chinese models followed suit because of, right, distillation, which we'll talk about a little bit here, in a couple of minutes. But you can't overlook Gemma because it puts Frontier capability on literally a laptop. Right? Because two years ago, to be able to run something like a g b d four o level model, right, which, rumors have been swirling that's it's a 2,000,000,000,000 parameter model. Right? You would need a small little data center to run something like that, you know, two ish years ago. Now you have these capabilities. So, you know, I've been lucky to talk to a lot of smart people in AI, and now you really have executives grappling with, well, should we be buying a bunch of, as an example, new MacBooks? Should we be, you know, buying a bunch of DGX sparks, for our employees and setting them up with twenty four seven always on agentic AI, right, to take advantage of these now local and powerful models that, well, you don't pay. Right? You download them once, you don't pay again, and they work and they can work while you sleep, like I said, because of, some of the new autonomous capabilities, from local agents that can run around the clock.
Jordan Wilson [00:17:04]:
So this has obviously led to, and I think, Google putting the pressure on the, the open source world with Gemma four. Like I said, it was 20 times more efficient in terms of what is it was able to achieve on the benchmarks in terms of size. Right? Because, when it came out, if you looked at the other Chinese models, it was about 10 to 20 times smaller in size. Right? So, you wouldn't have been able to, you know, use the best open source model pre Gemma four on a local machine, right, at least a consumer laptop that you can just go walk into the store and buy. Now you can. And open Chinese models are amazing, and I think they've been getting better and better and smaller and smaller and more and more efficient since Google's Gemma four, but you have to talk about the elephant in the room. That is these models are distilled. Right? We can say that all the big labs have said that, you know, are I don't know, maybe some of our, audience in China won't appreciate hearing that.
Jordan Wilson [00:18:11]:
But, I mean, Google, Anthropic, OpenAI have all accused China and have said they have proof, right, but the, the White House. So in April, the White House actually officially said that China was, using, kind of illegal, tactics to distill, USAI models to create cheaper domestic knockoffs. Right? So we got to the point that at least the White House said that they had enough, information or intel to make that declaration. So what is model distillation and, well, why does it matter? So the easiest way is, like, I can spend ten hours studying for a test. Right? Think back in the classroom. I can spend ten hours sitting for a test. Someone behind me can look over my shoulder and spend ten minutes and get the exact same answers. That's kinda like what model distillation is.
Jordan Wilson [00:19:07]:
Right? You have the big AI companies here in The US spending billions of dollars, right, on any single new, you know, model pre training as an example. And essentially, you have, certain actors in China who will use the API and, you know, different companies have come out with different levels of, proof and say, okay. Well, they're creating, you know, thousands of spoof accounts more or less. They're putting in all these inputs and training it on our model's outputs versus training it themselves. So, yeah, just kind of copying the homework. So what this has led to is China has been able to put out these open source models, really, technically, just pushing the frontier of open source by allegedly just copying the best US models out there. And what this has led to is, well, it's a crashing out at the bottom price of intelligence. So DeepSeq v four as an example.
Jordan Wilson [00:20:06]:
DeepSeq, one of those companies that many, of the AI labs here in The US have, accused of model distillation. DeepSeq v four Pro, one of their newer models, now list their price at 43¢, per million token inputs and 87¢ per million token outputs. That's, like, more than 25 times cheaper than the premium, you you know, closed source proprietary models. And that is the reality that a lot of boardrooms are looking at right now. Right? To make that math easy, it's like, okay. If we're spending a thousand dollars per month per employee on the API side, right, if you're or sorry. If you're spending, let's just say $40,000 a year, okay, we can be spending a thousand dollars a year if we switch over, to an open source model as an example. So that's here's what that actual leads that what that actually leads to.
Jordan Wilson [00:21:07]:
Right? Kind of the, the model distillation leads to more powerful, cheaper, open source models from China, and, well, it leads to people using them, but not always knowing the ramifications. Right? So, obviously, Google doing things the right way. But I think with these Chinese models, they've become increasingly popular even in the enterprise, which is tricky. And I don't think that most executives are fully understanding some of the consequences of using open source models. But this has led to essentially having a workforce of always on assistance, and they've shipped from expensive special projects to, well, that's just now the default operating model. So, you know, this has just allowed kind of these this new swarm of agentic AI that couldn't have really have existed before. Number one, the technology and the harnessing wasn't there. But number two, you take out, you you know, at least Gemma and the Chinese models, that it had been accused of distillation.
Jordan Wilson [00:22:15]:
And your your your options aside from those aren't really that good. Alright. We're gonna talk more here after we take a quick break for a word from our partners. Adobe just introduced an entirely new way to create, bringing the power and precision of its creative suite into one conversational experience. Meet Firefly AI assistant, now live in the Adobe Firefly app, the all in one creative AI studio. Powered by Adobe's creative agent, Firefly AI assistant lets you start with your vision, just describe what you want, and shape the outcome as it takes form with the assistant. The assistant orchestrates multi step workflows, drawing on 60 plus pro grade tools across Adobe Creative Cloud apps, including Photoshop, Illustrator Premiere, Lightroom Express, and more to help bring your ideas to life. You can also get started with creative skills, a growing library of pre built workflows for common creative tasks like batch editing photos, creating mood boards, portrait retouching, and creating social variations.
Jordan Wilson [00:23:18]:
Every step the assistant takes is visible, so you can refine, redirect, or take over at any time. You stay in the driver's seat as the creative director. Adobe Firefly AI assistant now in public beta. See it today at firefly.adobe.com. Adobe just introduced an entirely new way to create, bringing the power and precision of its creative suite into one conversational experience. Meet Firefly AI assistant, now live in the Adobe Firefly app, the all in one creative AI studio. Powered by Adobe's creative agent, Firefly AI assistant lets you start with your vision, just describe what you want, and shape the outcome as it takes form with the assistant. The assistant orchestrates multi step workflows, drawing on 60 plus pro grade tools across Adobe Creative Cloud apps, including Photoshop, Illustrator Premiere, Lightroom Express, and more to help bring your ideas to life.
Jordan Wilson [00:24:19]:
You can also get started with creative skills, a growing library of prebuilt workflows for common creative tasks, like batch editing photos, creating mood boards, portrait retouching, and creating social variations. Every step the assistant takes is visible, so you can refine, redirect, or take over at any time. You stay in the driver's seat as the creative director. Adobe Firefly AI assistant now in public beta. See it today at firefly.adobe.com. Aside from always on AI agents, what this open source movement has led to is, well, now enterprises may be moving away from having the one model fits all, solution. So now as an example, you might be able to put out an 100 agent swarm goes from, you know, 1,200 plus dollars on Opus to, well, maybe, like, 60 some dollars on DeepSeek. And now, essentially, you, can look at AI as more of a triage or a cat categorization of which models to use for which tasks.
Jordan Wilson [00:25:34]:
Right? Especially when you're talking about high volume operations. So things like when you're going through it in bulk, things like summarization, extraction, parsing PDFs, classification. Right? Now so many even large enterprise companies are no longer doing that on the back end using the frontier US, companies. Well, I mean, many still are, but you've already seen a big segment, of those companies move to these open source or, Chinese open source models. But if you're thinking right now, if you're like, wow, our bill is pretty high. Our API bill. Right? I'm not talking about on the front end, you know, the number of seats you have in chat GBT enterprise or, you know, in Gemini enterprise or anything like that. Right? I'm talking about back end, all of these special projects that you have running via the API.
Jordan Wilson [00:26:30]:
So if if if you're looking at your API bill and you're like, yeah. We're going through a lot. Or, you you know, hey. We're using, you know, Opus four seven and g p d five five to run our agents. Maybe we should be looking at, you know, Kimmy or Deepsea or, whatever it is. Before you do that, you have to know that there is a big trade off. Because just because a model is free or open source, or cheap ish to run via the API. Right? If you are, you know, running some of these open models, via the API.
Jordan Wilson [00:27:06]:
There's still, an expensive price to pay, and that price might be unknown at this point. But it can cost your company a lot more than maybe just using that closed proprietary AI would have cost would have cost you via the API. That's because using open source strips away all of that legal protection that you probably overlook or take for granted. What do I mean? Well, when you're using Anthropic, OpenAI, Microsoft, Who did I forget? Microsoft, OpenAI, Google, Anthropic. Right? When you're using those, enterprise offerings, you have a level of legal protection. Right? So as an example, if you use right. I'm not gonna go through all the, fine prints. Right? But the four companies at the enterprise level all offer, you know, some sort of essential, I won't call it insurance.
Jordan Wilson [00:28:05]:
Right? But think of it kinda like that. Right? Like, hey. If you use something produced by our systems and if you use it ethically and responsibly and with guardrails and it produces something that's not, you know, correct, there is some level of protection there. Right? Which you don't get that with open source models. Right? So as an example, you know, DeepSeek, you you know, they used, different MIT licenses, Apache two point o. Yeah. Essentially, there's no warranty or non infringement agreements. Alright? So for regulated work and customer facing output, you have to look at the trade off.
Jordan Wilson [00:28:43]:
Yeah. You might save $6.07, who knows? Maybe 8, figures by switching the bulk of some of your maybe agentic or bulk workloads. You're right. Especially if you're a fortune fortune five 100, fortune one 100 company. There's minimum $7.08 figures that you could, in theory, save by switching some of those heavier agentic, or, you know, parsing. You know, I know parsing is a big one, shifting some of those workflows to open models, but you lose that legal protection that maybe you've had to rely on it before. Maybe you haven't. But that one time that you would actually need it.
Jordan Wilson [00:29:26]:
And if you do switch over to open source, you have to understand those ramifications because at that point you're gonna actually be paying for it. So that gets us to the real question here as we get close to wrapping up, because I don't want my takeaway here to be don't use open models. They're not safe because that's not the takeaway. I think you need to start looking at your AI workflow like a triage. Right? At least when it comes to back end tasks. Right? Front end, I've always been a firm believer, and I still am today. You need to pick your AI operating system of choice, whether that's Copilot, Chad GPT, Claude, Gemini on the front end. And that's where you should move, especially your nontechnical people, should move the majority of their day to day knowledge work tasks should be happening on the front end there.
Jordan Wilson [00:30:22]:
But you still have a multitude of back end tasks. And I think you have to look at it like triaging in an emergency room. Right? You wouldn't send your top neurosurgeon in when someone's having an allergic reaction to honey. Right? You wouldn't do that. You would save that neurosurgeon for, well, someone that needs a neurosurgeon. And I think that there's so many companies that haven't gone through the basics of this. For the most part on the API side, they pick, well, one model and they say, alright, well, we have our AI operating system of choice. And then for everything else, as an example, we go to SONNET four six or we go to, you know, Gemini three one Flash or whatever that model may be.
Jordan Wilson [00:31:18]:
And maybe that's the right model. Maybe those companies have done their due diligence and have vetted out their different use cases and have priced it out, and maybe that's the right move. But maybe it's not. Because I know from experience and talking to a lot of people, a lot of companies just choose whatever is on the cutting edge. And they say, well, this is the best, so we're gonna pay for it because there is a push internally to use more AI, to use the best AI. We see all these new benchmarks. We wanna make sure that we're taking advantage of it. Well, is that neurosurgeon gonna be able to, you know, properly diagnose the allergic reaction to someone eating honey? Well, yeah, probably.
Jordan Wilson [00:32:00]:
But it's gonna cost you a lot more. So you need to think about sending those high volume, low stakes work, right, like summarization, research, content creation, maybe, to cheaper open source models that you can either run locally, or, you know, running via the API for just cost efficiency. If there's essentially, like, no legal ramifications if you get something wrong. Right? So if you're in a highly regulated sector, this is probably not the advice for you. Right? You shouldn't probably be using, you know, or just taking my advice on a whole lot of anything as truth. You always need to be vetting these things out for yourself. Right. But if if if if it is something relatively in a sector that's not highly regulated, where there's not a quote, unquote, a lot on the line, that's one of those instances where you need to say, can we shift some of our more expensive API workloads to an open source model? Or if you need to run sensitive private workflows on self hosted open models.
Jordan Wilson [00:33:03]:
That's another thing. I think that there's still, even to this day, even though I think there's plenty of reasons, you you know, one thing I always ask companies when they're like, oh, we don't well, we don't run this through AI. Right? Because it's sensitive data. And I'm like, okay. Well, do you have a cloud provider? And they're like, of course. It's like, okay. Well, it's the same thing, more or less. Right? As long as you take proper precautions, turn off model training, and all that, it's it's more or less using the same level grade of security that, you know, cloud, uses.
Jordan Wilson [00:33:31]:
Anyways, there are still some things that companies won't even put on the cloud. Right? Which I understand. But having these now, extremely powerful open source models and extremely efficient open source models, now you can start running those private workloads, workloads or workflows on prem, right, or, you know, self hosted that you can fully control. And then you can reserve those more premium, those more, high value, highly sensitive tasks, you know, that for those models that can reason and think and offer kind of that that that level of, security and legal support that you don't get if you opt for open models instead. So, you know, have a nice little chart here. So maybe when you look at the cheap open APIs, you look at simple task like summarization, extraction, or classification, for local self hosted open models. Private workflow's great for that, running agents locally, having more control. Right? And then for the premium closed models, which are the ones that a lot of people are using on the front end.
Jordan Wilson [00:34:48]:
On the back end, you should still be using these for a lot of reasons. Right? Those that, well, carry a lot of business value, hard tasks that require reasoning. Right, your final review. Maybe you do, you know, draft version either with a cheap API or a local self hosted open model, but or anything, you know, that requires customer facing output should probably be going on that premium closed models for that level of protection. Like I said, you cannot overlook the hidden trade off that these open licenses may disclaim warranty and non infringement where the enterprise offerings do usually include that I that IP, indemnification. So for regulated work, that protection in almost all cases justifies the premium that you pay. However, as we wrap up here, let me just quickly encapsulate all of this. Local models aren't going anywhere.
Jordan Wilson [00:35:52]:
Alright? And I actually think, especially as we, officially welcome in the era of models that can improve themselves and can create own versions, you know, smaller versions of themselves. Right? All the big companies have essentially said, you you know, have hinted, at RSI or, you know, the fact that our big models make smaller versions of themselves. I think that we're gonna not only see a continued trend toward, local open source models. I think we're gonna start seeing a lot of smaller models, for very specific use cases. It's something I've been, you know, predicting now for multiple years. We started to see it slowly. I think it is going to pick up steam now that we're starting to get some hints of recursive self improvement with these models. So your company has to be paying attention because this is a trend that is not going away.
Jordan Wilson [00:36:47]:
The open models are gonna become more and more capable. They're gonna become faster. They're gonna become more efficient, and the options are going to start to become even greater. Right? Not just great general purpose open models that can run on consumer hardware like Gemma four, but small open models for very specific tasks that can be highly valuable for your company. So you have to understand, the pros and the cons of these local models, when you might use a cheap API, and how this changes the agentic outlook for your company. So don't write them off just because you always wanna use the latest and the greatest. Yes. You should do that, but don't send the neurosurgeon to, you know, triage a a basic thing happening in the waiting room.
Jordan Wilson [00:37:40]:
Send the right model at the right time for the right purpose. So I hope this was helpful as we recapped open source AI one zero one as part of our start here series. If this was helpful, number one, make sure you subscribe to the podcast. I'd appreciate that. But then make sure you go to starthereseries.com. That's going to give you free access to our exclusive inner circle community. Right now, there's no other way to join except by going to starthereseries.com. So make sure you do that.
Jordan Wilson [00:38:07]:
Thank you for tuning in. I hope to see you back tomorrow and every day for more everyday AI. Thanks, y'all. Meet Firefly AI assistant now live in Adobe Firefly, the all in one creative AI studio. Just describe what you want to create in your own words, and the assistant handles the rest, orchestrating multistep workflows across Adobe Creative Cloud apps, including Photoshop, Premiere, Express, and more in one conversational interface. You direct the outcome while the assistant accelerates execution. Stay in control with the ability to step in and refine at any time. See it today at firefly.adobe.com.