Back

Montgomery v. Caribe Won’t Save Your Carrier Selection. Your Workflow Will.

Michelle McBride

This summer, brokerage leaders are watching one case: Montgomery v. Caribe. By early July, the Court will decide whether federal law shields brokers from negligent carrier selection suits. The stakes are real. Yet, so is the temptation to treat the ruling as a finish line, when even a clean win for brokers leaves the operational picture exactly where it sits today.

That's the uncomfortable truth: Montgomery won't fix your exposure. Your workflow will.

Why the Verdict Barely Matters

The legal question in Montgomery is narrow: when a broker picks a carrier and that carrier kills someone, can the victim sue the broker in state court, or does federal law block it? A ruling by July will settle it nationwide. Here's the case detail if you want it.

Now picture the best-case outcome for brokers: C.H. Robinson and Caribe win. State-court negligent selection suits go away. You still have a problem, because the market already moved without waiting for the Court.

Your insurer rewrote the carrier selection section of your renewal questionnaire. Your largest shippers added new indemnity language to their contracts. Plaintiff firms are building discovery requests around a tougher standard. None of that unwinds because nine justices in Washington side with brokers.

That's the disconnect the industry keeps missing. Montgomery decides one venue question. Your insurer, your shippers, and the next plaintiff's lawyer have already decided the operational one, and they've all landed in the same place. "We vetted them" used to be enough. What they want now is the timestamped record: what your team saw, when they saw it, and what they did about it.

Where Carrier Selection Breaks and What 2026 Keeps Adding

Here's the part the industry keeps glossing over. The what is settled. Every insurer, shipper, and plaintiff firm has converged on the same answer: a timestamped record of how the carrier got picked. That's the easy half of the conversation.

The how is the open question. How does that record get written on a floor running flat out, on the bookings most likely to end up in a deposition—the late-night recovery, the panicked re-source, the carrier nobody on the team has used before? 

The standard moved, the workflow didn't, and 2026 keeps pressing on exactly those moments.

When the Record Gets Made Under Pressure

Bad tenders don't show up when the desk is quiet. They show up when the customer is pressing for an update, the lane is heating, capacity is tight, and pickup is six hours out. The rep has a partial board, an inbox they haven't cleared in two hours, and a manager standing behind them asking why the load isn't covered. 

So they click the same three carriers they always use on that lane. They take the marginal MC because it's the only truck moving and the appointment won't shift. Then skip the deeper check because the next phone is already ringing and the load behind this one is starting to tip.

That sequence is your carrier selection record in discovery. Not the policy on the wiki. Not the training deck from the last all-hands. The decisions made inside the pressure window, by a tired rep, against a clock the shipper is setting.

Discipline and speed pull against each other under load, and no rep solves that by working harder. You can't policy your way out of decision latency. The fix has to live inside the workflow itself, not on a sticky note next to the monitor or another reminder bolted onto the TMS.

Four Months. Four New Pressures. One Already-Broken Window.

That broken booking moment, the late resource with a manager standing behind the rep, is exactly what 2026 keeps loading more weight onto. The regulatory calendar isn't dropping one new constraint the desk can absorb and move on. It's stacking four of them in four consecutive months, and every one lands on the same rep, on the same load, in the same window where the work was already breaking.  

Walk it month by month:

January 16. FMCSA Financial Responsibility Rule. 

Brokers now have to hold $75K in liquid security and replenish within seven days or lose operating authority. The carrier-side mirror of that rule is just as live: surety claims and authority pulls move faster, which means the carrier a rep cleared on Monday can be ineligible by Thursday. 

  • The Desk Effect: Vetting stops being a gate at onboarding and becomes a continuous obligation. The rep is now responsible for catching status changes between the last check and the next dispatch, with nothing added to the clock to do it.

February 26. SAFER Transport Act 

Sen. Todd Young's bill takes direct aim at fictitious pickups, double brokering, and hostage loads. Whether it passes this session or the next, insurers and shippers are already pricing in the higher identity-verification standard it signals. 

  • The Desk Effect: More verification steps on every first-time carrier, and disproportionately on the panicked resource where verification is most likely to get skipped. The booking most likely to end up in discovery is the booking the new identity expectations make heaviest.

March 16. Non-Domiciled CDL Rule. 

The rule eventually pulls up to 194,000 drivers out of eligibility. The desk effect: the pool of legitimate capacity shrinks, especially on the lanes that already run tight, which pushes the rep toward the marginal carrier they would have skipped on a calmer board. Capacity pressure and selection pressure feed each other, and they feed each other hardest in the exact window where the record is most likely to come apart.

  • The Desk Effect: The pool of legitimate capacity shrinks, especially on the lanes that already run tight, which pushes the rep toward the marginal carrier they would have skipped on a calmer board. Capacity pressure and selection pressure feed each other, and they feed each other hardest in the exact window where the record is most likely to come apart.

May. FMCSA Broker Transparency NPRM. 

The second NPRM lands with a 48-hour records-on-request standard. 

  • The Desk Effect: The audit window collapses. What used to be a discovery exercise nine months after the fact becomes a two-day production demand any party in the chain can trigger. There is no time to reconstruct a record that wasn't written when the load was booked.

Every Counterparty Wants the Same Thing

That same pressed rep is now being watched by five different audiences at once: insurers at renewal, shippers under the contract, FMCSA on a 48-hour clock, plaintiff counsel under oath, and the court of public opinion. All of them want the same artifact, and a workflow that depends on careful documentation when nobody has time to document carefully will lose every one of them. That’s the problem a process fix has to solve.

What a Defensible Carrier Selection Process Does

Strip away the consultant decks, and the fix is pretty concrete. Five elements show up wherever case law, plaintiffs’ attorneys, and insurers converge on carrier selection, and they share one thread. The audit trail can’t be a separate workflow. If it depends on rep discipline after the fact, it won’t survive a deposition.

1. A Minimum Vetting Floor That Runs on Every Load

Authority, insurance, and FMCSA safety data hit every single load. Trusted carriers don’t skip ahead, and post-pickup catch-up doesn’t fly. The floor runs especially when the rep is swamped

2. Automatic Checks at the Moment of Selection

MC/DOT, CSA scores, identity verification, and double-brokering signals fire when the rep picks the carrier, not 20 minutes after the load is already covered. Catching a chameleon carrier post-dispatch is just an expensive lesson learned the hard way.

3. Exception Codes With a Real Approval Path

When a carrier falls outside policy, the load routes to someone with actual authority to say yes or no. The conversation lives in the system. Defense counsel cannot subpoena a hallway “go ahead.”

4. A Timestamped Record That Generates Itself

The system records who checked what, when, and what the result was. Capture happens as a byproduct of booking the load. If documentation requires extra clicks after the fact, those clicks will not happen at a suboptimal time.

5. Captured Rationale for Why This Carrier

The why is the piece most brokerages skip. A defensible record shows what the rep checked and why this specific carrier fit this specific load: lane history, equipment match, and prior performance. That detail is what turns “we vetted them” into “here’s why we chose them.”

How Envoy AI’s Ellie Builds the Audit Trail Inside the Booking Workflow

Ellie is our agentic carrier rep, the Envoy AI Logistics Management Agent that does the booking instead of watching it get done. Not a dashboard your team has to feed. Nor another tool stapled to your TMS. We built her because the carrier rep job has gotten harder than any one person can carry, and the cost of getting it wrong has gotten steeper.

The bookings that end up in discovery fail for human reasons. Ellie is the junior rep who never gets tired, never skips a step when the floor gets loud, and never books the marginal MC because the next phone is already ringing. She sources, reaches out, negotiates, and verifies, and the audit trail comes out of how she covers the load.

  • Vetting Happens Before the First Email Goes Out: Every carrier Ellie sources gets run against authority, insurance, safety score, and identity signals before she ever picks up the phone. The vetting is the only way she’s allowed to start.

  • Every Outreach Lands in the Load Record on the Way Out: Calls, texts, and emails get timestamped and attached the moment they happen. Discovery looks like a clean ledger instead of three reps squinting at Outlook.

  • Rate Negotiation Lives Inside Your Guardrails: Ellie negotiates against the max pay, lane ceilings, and blacklists the brokerage already set. Anything outside those rules kicks back to the rep with the reason attached. Off-policy bookings get a human signature and a paper trail.

  • The Rep Approves, and That Approval Is the Decision on Record: Ellie does not book behind anyone’s back. The human stays in the loop, owns the call, and that approval click becomes the documented carrier selection decision.  

  • Every Load Compounds Into the Carrier Context Graph: Envoy AI’s proprietary semantic layer turns every booking, negotiation, and carrier interaction into a graph that the brokerage owns. The graph belongs to your floor, sits on your record, and gets sharper every shift.

The proof is on live floors. About 75% of freight inside Envoy accounts is being booked through Ellie. That means 75% of the carrier selection record is generating itself, fast enough for dispatch and structured enough for a deposition. The rep stops carrying two jobs at once. Ellie handles the receipt. The rep owns the relationship and the rate.

Trust Built This Industry. Proof Is What Comes Next.

For decades, this business ran on trust. You knew the carrier, your dispatcher knew their dispatcher, the lane behaved, and when someone needed paper you went and found it. That era is closing. Whatever the Supreme Court does with Montgomery, the operational bar already moved, and it only moves one direction from here. Toward proof.  

Call it what it is: proof-based execution. An agent does the work, a human stays in the loop, and the defensible record gets written while the load gets covered. Not a compliance feature bolted onto the old model. A new category of carrier procurement, built for an industry that no longer gets the benefit of the doubt.

The cost of standing still is quiet at first. Renewal premiums creep. Indemnity terms tighten on the shippers you need most. A lawsuit lands that should have been defensible and isn't, because the file couldn't tell the story. None of that hits in a quarter. It hits over a couple of renewal cycles, and by the time it shows up on the P&L the distance to the operators who moved early is too far to close.

This is where Ellie comes in, and not solely as a procurement tool. She is the execution layer the brokerage runs on, and she gets sharper every load. Every booking, every negotiation, every carrier interaction compounds into a record you own and a system that understands your lanes, your shippers, and your guardrails a little better than it did the load before. Not only does the operation move faster, but it gets smarter on its own, shift over shift, without anyone having to retrain it.

When the next plaintiff's attorney asks how this carrier got this freight, the answer should not be a story your ops team is reconstructing under oath nine months later. It should be a record Ellie already wrote.

Book a demo and see what Ellie looks like on your floor.

This summer, brokerage leaders are watching one case: Montgomery v. Caribe. By early July, the Court will decide whether federal law shields brokers from negligent carrier selection suits. The stakes are real. Yet, so is the temptation to treat the ruling as a finish line, when even a clean win for brokers leaves the operational picture exactly where it sits today.

That's the uncomfortable truth: Montgomery won't fix your exposure. Your workflow will.

Why the Verdict Barely Matters

The legal question in Montgomery is narrow: when a broker picks a carrier and that carrier kills someone, can the victim sue the broker in state court, or does federal law block it? A ruling by July will settle it nationwide. Here's the case detail if you want it.

Now picture the best-case outcome for brokers: C.H. Robinson and Caribe win. State-court negligent selection suits go away. You still have a problem, because the market already moved without waiting for the Court.

Your insurer rewrote the carrier selection section of your renewal questionnaire. Your largest shippers added new indemnity language to their contracts. Plaintiff firms are building discovery requests around a tougher standard. None of that unwinds because nine justices in Washington side with brokers.

That's the disconnect the industry keeps missing. Montgomery decides one venue question. Your insurer, your shippers, and the next plaintiff's lawyer have already decided the operational one, and they've all landed in the same place. "We vetted them" used to be enough. What they want now is the timestamped record: what your team saw, when they saw it, and what they did about it.

Where Carrier Selection Breaks and What 2026 Keeps Adding

Here's the part the industry keeps glossing over. The what is settled. Every insurer, shipper, and plaintiff firm has converged on the same answer: a timestamped record of how the carrier got picked. That's the easy half of the conversation.

The how is the open question. How does that record get written on a floor running flat out, on the bookings most likely to end up in a deposition—the late-night recovery, the panicked re-source, the carrier nobody on the team has used before? 

The standard moved, the workflow didn't, and 2026 keeps pressing on exactly those moments.

When the Record Gets Made Under Pressure

Bad tenders don't show up when the desk is quiet. They show up when the customer is pressing for an update, the lane is heating, capacity is tight, and pickup is six hours out. The rep has a partial board, an inbox they haven't cleared in two hours, and a manager standing behind them asking why the load isn't covered. 

So they click the same three carriers they always use on that lane. They take the marginal MC because it's the only truck moving and the appointment won't shift. Then skip the deeper check because the next phone is already ringing and the load behind this one is starting to tip.

That sequence is your carrier selection record in discovery. Not the policy on the wiki. Not the training deck from the last all-hands. The decisions made inside the pressure window, by a tired rep, against a clock the shipper is setting.

Discipline and speed pull against each other under load, and no rep solves that by working harder. You can't policy your way out of decision latency. The fix has to live inside the workflow itself, not on a sticky note next to the monitor or another reminder bolted onto the TMS.

Four Months. Four New Pressures. One Already-Broken Window.

That broken booking moment, the late resource with a manager standing behind the rep, is exactly what 2026 keeps loading more weight onto. The regulatory calendar isn't dropping one new constraint the desk can absorb and move on. It's stacking four of them in four consecutive months, and every one lands on the same rep, on the same load, in the same window where the work was already breaking.  

Walk it month by month:

January 16. FMCSA Financial Responsibility Rule. 

Brokers now have to hold $75K in liquid security and replenish within seven days or lose operating authority. The carrier-side mirror of that rule is just as live: surety claims and authority pulls move faster, which means the carrier a rep cleared on Monday can be ineligible by Thursday. 

  • The Desk Effect: Vetting stops being a gate at onboarding and becomes a continuous obligation. The rep is now responsible for catching status changes between the last check and the next dispatch, with nothing added to the clock to do it.

February 26. SAFER Transport Act 

Sen. Todd Young's bill takes direct aim at fictitious pickups, double brokering, and hostage loads. Whether it passes this session or the next, insurers and shippers are already pricing in the higher identity-verification standard it signals. 

  • The Desk Effect: More verification steps on every first-time carrier, and disproportionately on the panicked resource where verification is most likely to get skipped. The booking most likely to end up in discovery is the booking the new identity expectations make heaviest.

March 16. Non-Domiciled CDL Rule. 

The rule eventually pulls up to 194,000 drivers out of eligibility. The desk effect: the pool of legitimate capacity shrinks, especially on the lanes that already run tight, which pushes the rep toward the marginal carrier they would have skipped on a calmer board. Capacity pressure and selection pressure feed each other, and they feed each other hardest in the exact window where the record is most likely to come apart.

  • The Desk Effect: The pool of legitimate capacity shrinks, especially on the lanes that already run tight, which pushes the rep toward the marginal carrier they would have skipped on a calmer board. Capacity pressure and selection pressure feed each other, and they feed each other hardest in the exact window where the record is most likely to come apart.

May. FMCSA Broker Transparency NPRM. 

The second NPRM lands with a 48-hour records-on-request standard. 

  • The Desk Effect: The audit window collapses. What used to be a discovery exercise nine months after the fact becomes a two-day production demand any party in the chain can trigger. There is no time to reconstruct a record that wasn't written when the load was booked.

Every Counterparty Wants the Same Thing

That same pressed rep is now being watched by five different audiences at once: insurers at renewal, shippers under the contract, FMCSA on a 48-hour clock, plaintiff counsel under oath, and the court of public opinion. All of them want the same artifact, and a workflow that depends on careful documentation when nobody has time to document carefully will lose every one of them. That’s the problem a process fix has to solve.

What a Defensible Carrier Selection Process Does

Strip away the consultant decks, and the fix is pretty concrete. Five elements show up wherever case law, plaintiffs’ attorneys, and insurers converge on carrier selection, and they share one thread. The audit trail can’t be a separate workflow. If it depends on rep discipline after the fact, it won’t survive a deposition.

1. A Minimum Vetting Floor That Runs on Every Load

Authority, insurance, and FMCSA safety data hit every single load. Trusted carriers don’t skip ahead, and post-pickup catch-up doesn’t fly. The floor runs especially when the rep is swamped

2. Automatic Checks at the Moment of Selection

MC/DOT, CSA scores, identity verification, and double-brokering signals fire when the rep picks the carrier, not 20 minutes after the load is already covered. Catching a chameleon carrier post-dispatch is just an expensive lesson learned the hard way.

3. Exception Codes With a Real Approval Path

When a carrier falls outside policy, the load routes to someone with actual authority to say yes or no. The conversation lives in the system. Defense counsel cannot subpoena a hallway “go ahead.”

4. A Timestamped Record That Generates Itself

The system records who checked what, when, and what the result was. Capture happens as a byproduct of booking the load. If documentation requires extra clicks after the fact, those clicks will not happen at a suboptimal time.

5. Captured Rationale for Why This Carrier

The why is the piece most brokerages skip. A defensible record shows what the rep checked and why this specific carrier fit this specific load: lane history, equipment match, and prior performance. That detail is what turns “we vetted them” into “here’s why we chose them.”

How Envoy AI’s Ellie Builds the Audit Trail Inside the Booking Workflow

Ellie is our agentic carrier rep, the Envoy AI Logistics Management Agent that does the booking instead of watching it get done. Not a dashboard your team has to feed. Nor another tool stapled to your TMS. We built her because the carrier rep job has gotten harder than any one person can carry, and the cost of getting it wrong has gotten steeper.

The bookings that end up in discovery fail for human reasons. Ellie is the junior rep who never gets tired, never skips a step when the floor gets loud, and never books the marginal MC because the next phone is already ringing. She sources, reaches out, negotiates, and verifies, and the audit trail comes out of how she covers the load.

  • Vetting Happens Before the First Email Goes Out: Every carrier Ellie sources gets run against authority, insurance, safety score, and identity signals before she ever picks up the phone. The vetting is the only way she’s allowed to start.

  • Every Outreach Lands in the Load Record on the Way Out: Calls, texts, and emails get timestamped and attached the moment they happen. Discovery looks like a clean ledger instead of three reps squinting at Outlook.

  • Rate Negotiation Lives Inside Your Guardrails: Ellie negotiates against the max pay, lane ceilings, and blacklists the brokerage already set. Anything outside those rules kicks back to the rep with the reason attached. Off-policy bookings get a human signature and a paper trail.

  • The Rep Approves, and That Approval Is the Decision on Record: Ellie does not book behind anyone’s back. The human stays in the loop, owns the call, and that approval click becomes the documented carrier selection decision.  

  • Every Load Compounds Into the Carrier Context Graph: Envoy AI’s proprietary semantic layer turns every booking, negotiation, and carrier interaction into a graph that the brokerage owns. The graph belongs to your floor, sits on your record, and gets sharper every shift.

The proof is on live floors. About 75% of freight inside Envoy accounts is being booked through Ellie. That means 75% of the carrier selection record is generating itself, fast enough for dispatch and structured enough for a deposition. The rep stops carrying two jobs at once. Ellie handles the receipt. The rep owns the relationship and the rate.

Trust Built This Industry. Proof Is What Comes Next.

For decades, this business ran on trust. You knew the carrier, your dispatcher knew their dispatcher, the lane behaved, and when someone needed paper you went and found it. That era is closing. Whatever the Supreme Court does with Montgomery, the operational bar already moved, and it only moves one direction from here. Toward proof.  

Call it what it is: proof-based execution. An agent does the work, a human stays in the loop, and the defensible record gets written while the load gets covered. Not a compliance feature bolted onto the old model. A new category of carrier procurement, built for an industry that no longer gets the benefit of the doubt.

The cost of standing still is quiet at first. Renewal premiums creep. Indemnity terms tighten on the shippers you need most. A lawsuit lands that should have been defensible and isn't, because the file couldn't tell the story. None of that hits in a quarter. It hits over a couple of renewal cycles, and by the time it shows up on the P&L the distance to the operators who moved early is too far to close.

This is where Ellie comes in, and not solely as a procurement tool. She is the execution layer the brokerage runs on, and she gets sharper every load. Every booking, every negotiation, every carrier interaction compounds into a record you own and a system that understands your lanes, your shippers, and your guardrails a little better than it did the load before. Not only does the operation move faster, but it gets smarter on its own, shift over shift, without anyone having to retrain it.

When the next plaintiff's attorney asks how this carrier got this freight, the answer should not be a story your ops team is reconstructing under oath nine months later. It should be a record Ellie already wrote.

Book a demo and see what Ellie looks like on your floor.

Other articles you might like