NASDAQ: WYFI

Colocation/Data center services

We design, develop, and operate
data centers, through which we offer our hosting and colocation services. Our operational data centers meet the requirements of the Tier-3
standard, including N+1 redundancy architecture, concurrent maintainability, uninterruptible power supply, advanced and highly reliable
cooling systems, strict monitoring and management systems, 99.982% uptime and no more than 1.6 hours of downtime annually, service organization
control, SOC 2 Type 2, differentiated software supporting AI workloads, high density and robust bandwidth, and infrastructure to support
AI workloads.

Based on their collective
industry experience, our data center team is adept at bringing new sites online on an accelerated timeline. We are aggressively pursuing
our development pipeline and intend to achieve an estimated 76 MW (gross) of total data center capacity by the end of the fourth quarter
of 2026, a target that is underpinned by assets including our MTL-2, MTL-3, and NC-1 facilities. As of December 31, 2025, our pipeline
of potential data center projects represents approximately 1,500 MW (gross) under management review. We follow a disciplined process prioritizing
projects that are backed by customer lease commitments. In select cases, we may pursue early-stage acquisitions based on strong customer
demand signals and defined commercialization pathways. Accordingly, the foregoing timelines and capacities are subject to change based
on many factors, many of which are outside of our control.

1

We use a well-defined set
of criteria to select our data center sites. We typically target sites with proximity to metro areas and partial infrastructure in place,
where we are retrofitting rather than developing greenfield projects. Metropolitan areas are positioned for low-latency to address long-term,
specialized AI computer inference needs, and smaller sites reduce risks. A retrofit entails sourcing and acquiring an existing industrial
building with underutilized, in-place power connectivity. The period of time from when a site is purchased until construction can begin
varies from location to location depending upon, among other things, obtaining required permits and the availability of construction supplies
and contractors. Average build time for retrofits is intended to be approximately six months from commencement of construction, which
we believe is approximately one-third to one-half of the industry average development timeline for greenfield projects. This average building
time is based upon senior management’s experience at Enovum prior to its acquisition by the Company, as well as their experience
prior to Enovum. We also prioritize sites offering opportunities to increase site power over time, enabling our data centers to grow with
customer demand. In addition, we selectively target certain larger opportunities with 50 MW (gross) of power or more, subject to customer
demand, to drive AI-driven compute super-clusters. Finally, we prioritize sites powered by sustainable, green energy sources and locked-in
power when available. Additionally, to enhance sustainability of certain of our data center projects, we are undertaking heat repurposing
projects in connection with sustainability and commercial and residential projects.

We acquired Enovum on October 11, 2024. The transaction included the
lease of MTL-1, our 4 MW (gross) Tier-3 high-performance computing (“HPC”) data center in Montreal, Canada, which was fully
operational and fully leased to customers at the time of acquisition.

On December 27, 2024, we
acquired the real estate and building for a build-to-suit 5 MW (gross) Tier-3 data center expansion project near Montreal, Canada which
we refer to as MTL-2. MTL-2, a 160,000 square foot site that was previously used as an encapsulation manufacturing facility, is located
in Pointe-Claire, Quebec. We initially funded the purchase of CAD 33.5 million (approximately $23.3 million) with cash on hand. We expected
to invest approximately $23.6 million to develop the site to Tier-3 standards with an initial load of 5 MW (gross). However, we have prioritized
other builds and preserved capital for more time sensitive projects.

On April 11, 2025, we entered into a lease for a new data center site
in Saint-Jerome, Quebec, a suburb of Montreal, MTL-3. The MTL-3 facility spans approximately 202,000 square feet on 7.7 acres and is being
developed into a 7 MW (gross) Tier-3 data center. It will support current contracted capacity, with Cerebras (5 MW IT Load), with future
expansion potential subject to utility approvals. The transaction was executed under a lease-to-own structure, which includes a fixed-price
purchase option of CAD 24.2 million (approximately $17.3 million) exercisable by December 2025. The lease term is 20 years, with two 5-year
extensions at the Company’s option. In December 2025, we became reasonably certain to exercise the purchase option and notified
the lessor of our intent to exercise the purchase option. We have 90 days to complete the purchase, after which the purchase option will
expire. The option was exercised on January 14, 2026 and the purchase of MTL-3 is expected to close during the second quarter of 2026.
The facility has been retrofitted to Tier-3 standards and was completed and operational in November 2025. The site has commenced billing
Cerebras as of November 1, 2025, in the amount of CAD 1.4 million (approximately 979 thousand USD) monthly for the duration of the five-year
contract.

On May 20, 2025, we completed
the purchase of a former industrial/manufacturing building from UMI. Pursuant to the Purchase Agreement we agreed to purchase from UMI,
an industrial/manufacturing building together with the underlying land located in Madison, North Carolina, which we refer to as “NC-1”,
as well as certain machinery and equipment located thereon for a cash purchase price of $45 million. The purchase price will increase
by (i) $8 million, if Duke Energy actually provides, or provides an Electric Services Agreement providing for, at least 99 MW (gross)
within two years of May 20, 2025, or (ii) $5 million, if Duke Energy actually provides, or provides an Electric Services Agreement providing
for, at least 99 MW (gross) more than two years but less than three years after May 20, 2025. Additionally, the purchase price will increase
by an additional $200,000 per MW over 99 MW (gross) up to a maximum of $5 million if at least 99 MW (gross) are actually delivered, or
Duke Energy provides an Electric Services Agreement for the provision of at least 99 MW (gross), within four years of May 20, 2025. Separately,
the Company entered into a Capacity Agreement with Duke Energy pursuant to which Duke Energy agreed to use commercially reasonable efforts
to achieve 24 MW (gross) of service to NC-1 by September 1, 2025, 40 MW (gross) by April 1, 2026, and 99 MW (gross) within four years
of May 16, 2025. Management believes based upon its review of the site and a Duke Energy preliminary transmission study, that NC-1 may
receive and support up to 200 MW (gross) of total electrical supply over an extended period of time, subject to infrastructure upgrades,
such as developing new substations and other conditions. On August 4, 2025, Enovum NC-1 Bidco LLC, a subsidiary of the Company, entered
into an Assignment and Assumption Agreement with Unifi Manufacturing and Duke Energy Carolinas, LLC, pursuant to which Enovum assumed
Unifi’s rights and obligations under certain electric service agreements for facilities located in North Carolina. Duke Energy
consented to the assignment. Refer to Note 17. Commitments and contingencies for further detail.

2

RBC
Facility Agreement Executed on June 18, 2025

On June 18, 2025, we entered
into the Credit Facility with RBC. The Credit Facility provides for an aggregate of up to approximately CAD 60 million (approximately
$43.8 million) of financing. The proceeds are to be used primarily to refinance the buildout of MTL-2 as well as $5.8 million of revolving
term financing (the “Revolver”). The Credit Facility is non-recourse to the Company. We entered into a three-year USD $18.5
million non-revolving lease facility to finance equipment costs and building improvements to build out the site. The lease facility provides
for straight-line amortization of six years and capital moratorium of six months after disbursement is complete. RBC may cancel any unutilized
portion of the Credit Facility after March 31, 2026. The interest rate is fixed based on the rental rate determined by RBC for the three-year
term of the lease.

As part of the Credit Facility,
we entered into a three-year $19.6 million non-revolving real estate term loan facility. The purpose of this facility is to refinance
the Company’s purchase of MTL-2. The interest rate of the real estate term loan facility will be determined at the time of borrowing,
or a floating interest rate ranging from RBP plus 0.75% to CORRA (“Canadian Overnight Repo Rate Average”) plus 250 bps. Payment
of principal and interest is due 30 days after drawdown and is repayable in full on the last day of the three-year term.

The Revolver is being provided by RBC by way of Letters of Credit and
Letters of Guaranty with fees to be determined on a transaction by transaction basis. This facility will be available for the 36 month
term subject to the issuance of the EDC (Export and Development Canada) Performance Security Guaranty in the amount of $5.8 million and
other related supporting documents. We agreed to certain financial covenants included maintaining on a combined basis between MTL-1 and
MTL-2: fixed charge coverage of not less than 1.20:1 and a ratio of Net Funded Debt to EBITDA of not greater than 4.25:1 and decreasing
to 3.50:1 from December 31, 2027.

In November 2025, our wholly
owned subsidiary, Enovum NC-1 Bidco, LLC, entered into the Services Agreement with Nscale Services US Inc. and Nscale Global Holdings
Limited (collectively, “Nscale”) for the provision of colocation and related services at our NC-1 facility. The agreement
represents a significant commercial milestone for our high-density data center platform and provides long-term contracted revenue visibility.
The initial Service Order pursuant to the Services Agreement represents approximately $865 million in total contracted revenue over a
10-year term, inclusive of contractual annual rate escalators and non-recurring installation services (“NRCs”). Electricity
and certain other operating costs are structured as pass-through charges to Nscale. Billing for the 40 MW phase is expected to commence
in June 2026, subject to completion of construction and commissioning. As a result, we expect revenue contribution from this agreement
to begin during the second quarter of 2026 as the facility reaches full contractual capacity.

3

Cloud Services

We provide specialized cloud services to support
generative AI workstreams, especially training and inference, emphasizing cost-effective utility and tailor-made solutions for each client.
We are an authorized NVIDIA Preferred Partner through the NPN, an authorized partner with SuperMicro Computer Inc.®, an authorized
CSP with Dell (through Dell’s exclusive distributor in Iceland, Advania), an official partnership with Hewlett Packard Enterprise
and a commercial relationship with QCT. Based on Management’s knowledge of the industry, we are proud to be among the first service
providers to offer H200, B200, and GB200 servers. We provide a high-standard service lease with an Uptime percentage> 99.5%.

We expect to leverage a global network of data
centers for hosting capacity for our GPU business, in many instances, by negotiating with third-party providers to seamlessly integrate
our cloud services at data centers across key regions in Europe, North America and Asia. Our initial data center partnership through which
we lease capacity is at Blönduós Campus, Iceland, offering a world-class operations team with certified technicians and reliable
engineers. The facility has a 45 kW rack density and 6 MW (gross) total capacity. We have executed contracts for 5.5 MW IT load at the
data center. The center’s energy source is 100% renewable energy, mainly from Blanda Hydro PowerStation, the winner of an IHA Blue
Planet Award in 2017. In addition, we have leased additional capacity to install our data center in Atlanta, Georgia, USA to expand
our cloud services offering. The capacity leases commenced in February 2026. We also intend to lease additional capacity to expand our
cloud services offering.

In April 2025, we received our first shipment
of NVIDIA GB200 NVL72 GPU server powered NVIDIA GB200 Grace Blackwell Superchips., from Quanta Cloud Technology, a global leading Original
Design Manufacturer (ODM). We believe that support with proof of concept (POC) access from Quanta will enable us to meet and exceed expectations
around delivery and timeline, performance and reliability.

The following summaries reflect selected GPU cloud
service agreements that we consider to be material or representative. We have entered into additional agreements that are not individually
material and are not included below.

On October 23, 2023, Bit Digital announced that
it had commenced AI operations by signing a binding term sheet with a customer (the “Initial Customer”) to support the customer’s
GPU workloads. On December 12, 2023, we finalized a Master Services and Lease Agreement (“MSA”), as amended, with our Initial
Customer for the provision of cloud services from a total of 2,048 GPUs over a three-year period. To finance this operation, we entered
into a sale-leaseback agreement with a third party, selling 96 AI servers (equivalent to 768 GPUs) and leasing them back for three years.
The total contract value with the Initial Customer for the aggregated 2,048 GPUs was estimated to be worth more than $50 million of annualized
revenue. On January 22, 2024, approximately 192 servers (equivalent to 1,536 GPUs) were deployed at a specialized data center and began
generating revenue, and subsequently on February 2, 2024, approximately an additional 64 servers (equivalent to 512 GPUs) also started
to generate revenue.

In the second quarter of 2024, we finalized an
agreement to supply our Initial Customer with an additional 2,048 GPUs over a three-year period. To finance this operation, we entered
into a sale-leaseback agreement with a third party, agreeing to sell 128 AI servers (equivalent to 1,024 GPUs) and leasing them back for
three years. In late July, at the customer’s request, we agreed with the customer to temporarily delay the purchase order so the
customer could evaluate an upgrade to newer generation Nvidia GPUs. Consequently, the Company and manufacturer postponed the purchase
order. In early August, the customer made a non-refundable prepayment of $30.0 million for the services to be rendered under this agreement.

In January 2025, the Company entered into a new
agreement to supply its Initial Customer with an additional 464 GPUs for a period of 18 months. This new agreement replaces the prior
agreement whereby the Company was to provide the customer with an incremental 2,048 H100 GPUs. The contract represents approximately $15
million of annualized revenue and features a two-month prepayment from the customer. The customer elected to defer the commencement date
until August 20, 2025, which is the latest allowable date under the agreement. Deployment commenced on August 20, 2025, using the Company’s
inventory of B200 GPUs.

In October 2025, the Company’s existing
parent guaranty arrangement with the Initial Customer was scheduled to expire. Beginning in November 2025, the customer will provide a
service deposit to the Company in lieu of the parent guaranty. The deposit will be funded through fifteen consecutive monthly payments
of approximately $0.24 million each, totaling $3.6 million, payable from November 2025 through January 2027. The deposit will serve as
security for the customer’s performance obligations under the amended service agreements. Each monthly payment is expected to be
invoiced on the first day of the month and paid within thirty days. The Company will be required to return the deposit in cash upon termination
or expiration of the service agreements, provided that all obligations have been fully satisfied and no payment defaults or material breaches
exist.

As of the report date, the Company and the Initial Customer are engaged
in discussions regarding a potential resolution of the existing service agreements following the agreed pause of services. No definitive
termination or settlement agreement has been executed. In connection with these discussions, the parties are negotiating the treatment
of the remaining non-refundable prepayment, service deposit, outstanding receivables, and a potential early termination fee, which the
Company believes would be equal to 40% of the fees that would have accrued for services during the remainder of the term of the MSA and
applicable purchase orders. Following the service pause, the Company has redeployed the GPUs previously allocated to the Initial Customer
to three other customers and continues to evaluate the related financial and operational implications. There can be no assurance as to
the timing, terms, or final outcome of these discussions.

4

On November 6, 2024, we entered into a Master
Services Agreement (“MSA”) with a minimum purchase commitment of 16 GPUs, along with an associated purchase order, from a
new customer. The purchase order provides for services utilizing a total of 16 H200 GPUs over a minimum of a six-month period, representing
total contracted value of approximately $160,000 for the term. The deployment commenced on November 7, 2024, using the Company’s
existing inventory of H200 GPUs. The service under the purchase order concluded in May 2025. Between May 2025 and September 2025, the
Company signed six additional agreements on a month-to-month basis for a total of 88 H200 GPUs, of which 80 remained in deployment as
of the date of this report.

In February 2026, we entered
into another service order with the customer to provide services utilizing a total of 10 H200 GPU servers. The service order has an initial
term of 14 months beginning on the services commencement date. The service order represents an aggregate revenue opportunity of approximately
$1.3 million. The deployment and revenue generation began in March 2026.

In March 2026, we entered into another service order with the customer
to provide services utilizing a total of 256 H100 GPU servers. The service order has an initial term of 24 months beginning on the services
commencement date, with an option to renew for an additional twelve months. The service order represents an aggregate revenue opportunity
of approximately $50.2 million. The deployment and revenue generation is expected to begin during the second quarter of 2026.

On November 14, 2024, we entered into a Terms
of Supply and Service Level Agreement (together, the “Agreement”) and an Order Form with a new customer. The order form provides
for services utilizing a total of 64 H200 GPUs on a month-to-month basis, which either party may terminate upon at least 14 days’
written notice prior to any renewal date. It represents annual revenue of approximately $1.2 million. The deployment commenced and revenue
generation began on November 15, 2024, using the Company’s existing inventory of H200 GPUs. The service under the purchase order
concluded in December 2024.

On December 30, 2024, we entered into a Master
Services Agreement (“MSA”) with an AI Compute Fund managed by DNA Holdings Venture Inc. (“DNA Fund”). The MSA
had a minimum purchase commitment of 32 GPUs, along with an associated purchase order. The purchase order provides for services utilizing
a total of 576 H200 GPUs over a 25-month period and terminable by either party upon at least 90 days’ written notice prior to any
renewal date. Concurrently, we placed a purchase order for 130 H200 servers for approximately $30 million. The deployment commenced in
February 2025.

In April 2025, the Company signed two additional
cloud services agreements with DNA Fund. The first agreement includes 104 NVIDIA H200 GPUs under a 23-month term and was deployed in May
2025. The second agreement includes 512 H200 GPUs under a 24-month term and was deployed in July 2025. With these additions, DNA Fund’s
total contracted deployment increased to 1,192 GPUs.

In November 2025, we terminated the MSA and all related purchase orders
with DNA Fund in accordance with the terms of the contract. At the time of termination, we had approximately $7.3 million in outstanding
accounts receivable. Pursuant to the termination agreement, the customer agreed to repay the outstanding balance. As of the date of this
report, we have collected $2.1 million of the outstanding amount.

On January 6, 2025, we entered into a Master Services
Agreement (“MSA”) with a minimum purchase commitment of 32 GPUs, along with an associated purchase order, from a new customer.
The purchase order provided for services utilizing a total of 32 H200 GPUs over a minimum of six-month period, representing total revenue
of approximately $300,000 for the term. The deployment commenced and revenue generation began on January 8, 2025, using the Company’s
existing inventory of H200 GPUs. The service under the purchase order concluded in April 2025 following a change in the customer’s
ownership, and the customer paid the remaining contract value as an early termination penalty.

5

In January 2025, we entered into a Master Services
Agreement (“MSA”), along with two associated purchase orders, from a new customer. The purchase orders provide for services
utilizing a total of 24 H200 GPUs over a minimum 12-month period, representing total revenue of approximately $450,000 for the term. The
deployment commenced and revenue generation began on January 27, 2025, using the Company’s existing inventory of H200 GPUs. The
service under the purchase order concluded in March 2025 after the customer ceased operations.

On January 30, 2025, we entered into a Master
Services Agreement (“MSA”) with a minimum purchase commitment of 40 GPUs, along with an associated purchase order, from a
new customer. The purchase orders provide for services utilizing a total of 40 H200 GPUs over a minimum of 12 month period, representing
total revenue of approximately $750,000 for the term. The deployment commenced and revenue generation began on January 24, 2025, using
the Company’s existing inventory of H200 GPUs. In October 2025, the purchase order was amended to reduce the number of H200 GPUs
from 40 to 8 and to extend the term of service through May 2027. Between April and July 2025, the Company signed four additional agreements
on a month-to-month basis for a total of 184 H200 GPUs, which were terminated in August 2025.

In March 2025, we entered a strategic partnership
with Shadeform, Inc., the premier multi-cloud GPU marketplaces, to bring on-demand NVIDIA B200 GPUs to customers beginning in May 2025.

In August and September 2025, we entered into
three service orders with a new customer. Each order form provides for services utilizing a total of 64 B200 GPUs on a weekly basis, which
either party may terminate by not extending it with mutual written agreement. In September, the customer renewed one order form for an
additional week for services utilizing a total of 64 B200 GPUs. As of the reporting date, no additional renewals have occurred.

In September 2025, we entered
into a service order with a new customer, which provides services utilizing a total of 16 B200 GPUs on a monthly basis, automatically
renewing for an additional one month period unless and until otherwise terminated upon at least seven days’ prior written notice.
The deployment commenced and revenue generation began on September 23, 2025. The agreement was not renewed after the initial term.

In October 2025, we entered into a service order
with a new customer to provide services utilizing a total of 48 H200 GPUs. The service order had an initial term of 36 months. The deployment
commenced and revenue generation began on October 21, 2025. The contract was terminated in December 2025.

In October 2025, we entered
into a two-week service order with a new customer to provide services utilizing a total of 72 B200 GPUs. In January 2026, we entered into
an additional two-week service order with this customer for 72 B200 GPUs.

In February 2026, we entered
into a further service order with this customer to provide services utilizing a total of 384 B200 GPUs. This service order has an initial
term of 24 months commencing on the service commencement date, after which it will automatically renew for successive one-month periods
unless terminated by either party. The service order represents an aggregate revenue opportunity of approximately $18.1 million. Deployment
and revenue generation commenced on January 27, 2026.

In November 2025, we entered into a service order with a new customer
to provide services utilizing a total of 128 B200 GPUs. The service order has an initial term of 12 months, representing total contracted
value of approximately $3.0 million, after which it automatically renews for successive one-month periods unless terminated by either
party. Deployment and revenue generation began on December 1, 2025.

In February 2026, we entered
into a service order with a new customer to provide services utilizing a total of 256 GPUs. The service order has an initial term of 12
months beginning on the services commencement date, after which it automatically renews for successive one-month periods unless terminated
by either party. The deployment and revenue generation began on February 1, 2026.

In March 2026, we entered into a service order with a new customer
to provide services utilizing a total of 72 GB200 GPUs. The service order has an initial term of 12 months beginning on the services commencement
date, after which it automatically renews for successive one-month periods unless terminated by either party. The deployment and revenue
generation began on March 7, 2026.

In August 2024, we executed a binding term sheet
with Boosteroid Inc. (“Boosteroid”), a global cloud gaming provider pursuant to which, we finalized initial orders of 489
GPUs, projected to generate approximately $7.9 million in contracted value in the aggregate through November 2029. The GPUs were delivered
to respective data centers across the U.S. and Europe and began earning fees in November 2024. On October 9, 2024, we executed a Master
Services and Lease Agreement (the “MSA”) with Boosteroid, pursuant to which Boosteroid may, from time to time, lease certain
equipment, including GPUs, from the Company upon delivery of a purchase order. The MSA provides the general terms and conditions for such
equipment leases. Pursuant to the MSA, we are granted a right of first refusal with respect to the next 5,000 servers that Boosteroid
leases during the term of the MSA. The MSA provides Boosteroid with the option to expand in increments of 100 servers, up to 50,000 servers,
representing a potential contract value of approximately $700 million over the five-year term assuming Boosteroid utilizes the GPUs and
services at full capacity for the duration of the contract. Expansion depends upon the internal development roadmap of Boosteroid, Boosteroid
has full discretion to decide when and the quantity to pursue separate source orders (for GPU servers) under the MSA. In the third quarter
of 2025, the Company finalized additional purchase orders for 302, 120, and 279 GPUs, totaling approximately $10.4 million in contracted
value over a five-year term.

6

Reorganization, Initial Public Offering, and Relationship with Bit
Digital

We were incorporated by Bit
Digital as a Cayman Islands exempted company on August 15, 2024 under the name Celer, Inc., as a holding company for the HPC Business.
We changed our name to WhiteFiber, Inc. on October 17, 2024.

On August 6, 2025, we issued
27,043,749 ordinary shares, par value $0.01 per share (our “Ordinary Shares”, and such shares, the “Contribution Shares”),
to Bit Digital pursuant to the terms of a Section 351 Contribution Agreement (the “Contribution Agreement”) entered into with
Bit Digital on July 30, 2025. Pursuant to the Contribution Agreement, Bit Digital contributed its HPC Business through the transfer of
100% of the capital shares of its cloud services subsidiary, WhiteFiber AI, Inc. and its wholly-owned subsidiaries WhiteFiber HPC, Inc.,
WhiteFiber Canada, Inc., WhiteFiber Japan G.K. and WhiteFiber Iceland, ehf, to us, upon the effectiveness of the registration statement
filed in connection with our IPO and prior to the consummation of the IPO, in exchange for the Contribution Shares. We refer to this transaction
as the “Reorganization”. WhiteFiber AI became a wholly-owned subsidiary of WhiteFiber, Inc. and Bit Digital became the direct
shareholder of WhiteFiber after the Reorganization.

On August 8, 2025, we completed our initial public offering ( “IPO”)
of 9,375,000 Ordinary Shares, at a public offering price of $17.00 per share. The gross proceeds to the Company from the IPO were approximately
$159.4 million, before deducting underwriting discounts and commissions and offering expenses of $12.0 million. On September 2, 2025,
B. Riley Securities, Inc. and Needham & Company, LLC, as representatives of the several underwriters of the IPO, fully exercised their
option to purchase an additional 1,406,250 Ordinary Shares at the public offering price of $17.00 per share, resulting in additional gross
proceeds to the Company of approximately $23.9 million.

After giving effect to the
IPO and the full exercise by the underwriters of their over-allotment option, Bit Digital held approximately 71.5% of our issued and outstanding
Ordinary Shares.

Following our IPO, certain
of our directors, executive officers and other members of senior management continue to serve as directors, officers and employees of
Bit Digital. We have added additional executive officers and senior management to our senior executive team apart from those serving as
officers and employees of Bit Digital. We have assembled a senior operating team with approximately 15 years of experience on average
for each individual in the data center and cloud services industries. We also appointed additional independent directors upon the commencement
of trading of our Ordinary Shares on Nasdaq.

Industry Overview

We compete in the large and
rapidly growing data center and cloud services markets. The data center market refers to the industry dedicated to designing, building,
and managing data centers, essential for storing, processing, and managing vast amounts of digital information, including for AI and ML
applications. These centers house servers, networking equipment, power and cooling and storage systems, ensuring seamless and secure data
operation.

The cloud services market
represents the transmission of computer services, including storage, analytics, databases, networking, and intelligence, via the internet
(referred to as “the cloud”). Our cloud services business is a NeoCloud provider, with differentiated services supporting
AI workloads and specializing in deploying optimized infrastructures that include not just GPU fleets but also network accelerators, high-speed storage,
software solutions, advanced orchestration tools, high-performance interconnects, edge computing capabilities, innovative cooling
solutions, security and compliance features, 24/7 managed services, and hybrid cloud integrations. This infrastructure is essential for
managing the massive data and computing demands of AI and ML training and inference tasks.

7

It is standard for GPU hardware
clusters, used in the rendering of cloud services, to be remotely housed in data centers, making these two segments highly related and
synergistic. Specially designed data centers host the hardware needed to provide HPC cloud services, and these are commonly referred to
as HPC data centers. According to experts at Schneider Electric, the industry is shifting away from traditional data centers that typically
hosted 10kW racks to newer 100kW racks that support HPC needs. We only design and develop HPC data centers, giving us a competitive advantage
over operators of traditional data center capacity.

Colocation/Data Center Services

According to McKinsey &
Company, power demand for data centers in the U.S. — driven by the need for digital and AI capabilities — is
expected to reach 298 gigawatts by 2030, up from 60 gigawatts in 2024. McKinsey estimates that data centers will amount to 11.7% of total
U.S. power demand in 2030. Based on our knowledge of the industry, we believe we are leaders in the markets that are critical for
these capabilities, and, as a result, we believe we are well positioned to benefit from the growth of this sector.

According to Prescient and
Strategic Intelligence, Data-Center Market Size and Analysis, Trends, Drivers, Competitive Landscape and Forecast (2024-2030), the
global data center market was valued at $342 billion in 2023 and is anticipated to reach $622.4 billion by 2030, expanding at
a CAGR of 10.5% during 2024-2030. As of 2023, the data center services segment, which provides a range of offerings including consulting,
maintenance, and management to optimize the performance and reliability of data center infrastructure, represented about 34.2% of the
industry. The solutions segment, which refers to the technological offerings that fulfill key functions within data center infrastructure,
comprised about 65.8% of the industry. The solutions segment represents both hardware and software elements, encompassing servers, storage
systems, advanced power systems, infrastructure networking equipment, and management software.

The data center market is
propelled by growing demands for cloud computing services, big data analytics, and digital transformation. According to McKinsey &
Company, by 2030, 70% of data centers expected to be developed will be for advanced AI, and 92% of companies plan to increase AI investments
across the board. Key contributors to this dynamic landscape are technology firms, real estate developers, and service providers. Together,
they play a vital role in the ongoing evolution of data center infrastructure, adapting to meet the expanding requirements of the digital
era and ensuring the seamless operation of critical digital services across various industries. The increasing global demand for cloud
computing, where we compete as a cloud services provider, is a major driver of data centers, underlining the inherent synergies between
our businesses.

Cloud Services

The global cloud AI infrastructure
market is forecasted to grow from $60.5 billion in 2024 to $363.4 billion in 2030, a compound annual growth rate of approximately
35%, according to research published by Mordor Intelligence. The major factors driving cloud services market growth are increasing digital
transformation across businesses, growing internet and mobile device adoption across the globe, and, most notably, increasing usage of
large data sets. Moreover, cloud services are favored by customers due to low capital investment, resource scalability, and a high degree
of accessibility. The rise of the system of connected devices, edge computing, 5G, and real-time analytics driven by AI and ML is
anticipated to increase the market value of cloud technology across different businesses.

8

Sustainability

Sustainability and energy
efficiency are increasingly important considerations for the data center and cloud services markets due to high energy consumption and
carbon emissions. The sustainability movement for data centers is driven by organizations’ environmental, social and governance
commitments and the rise of laws and regulations supporting sustainability. For example, the Paris Accord, which was entered into force
on November 4, 2016, is currently in effect across 174 countries apart from the United States and aims to curb long-term global
warming. Our facilities in Quebec and Iceland benefit from clean, hydroelectric power generation, and we will seek to offer comprehensive
HPC data center and cloud services solutions while prioritizing sustainability and energy efficiency.

Our Competitive Strengths

Robust customer acquisition capabilities
within large and growing total addressable markets.

As described above, both
the data centers and cloud services sectors are currently undergoing a well-documented surge in demand, driven in part by the proliferation
of AI models, agents and applications, and constraints on available supply. This growth is demonstrated by our ability to pre-sign end
users prior to committing capital for expansions, both for new data center sites and for GPU procurement. Our HPC data center business
enjoys significant embedded demand from existing customers. We expect that this demand will also enable us to extend the average duration
of our customer contracts. Finally, across our business units, we have invested in building robust sales, marketing, and customer acquisition
teams, and in developing our go-to-market strategy, including initiatives aimed at acquiring new customers through strategic partners,
outbound outreach, industry events, and third-party referral agents.

Confidential pipeline of WhiteFiber data
center opportunities with scalable, differentiated development capabilities.

Based on management’s prior experience, we believe our WhiteFiber
data center team are experts at sourcing attractive new development opportunities. Our management has previously identified and executed
on high-value data center sites, often in competitive or constrained power markets, such as Montreal, Quebec and Madison, North Carolina.
Their internal capabilities span site selection, power procurement, permitting and scalable design. These are complimented by deep industry
relationships that provide early access to off-market or strategic opportunities. Thanks to this capability, as of December 31, 2025,
our pipeline of potential data centers represents approximately 1,500 MW (gross) projects under management’s review.

Our opportunity pipeline
contains multiple sites for retrofit or brownfields. However, certain opportunities could from time to time necessitate a greenfield project
that entails a ground-up development. Our development team members are specialists at retrofits, having successfully delivered 75
MW (gross) of retrofit projects over their careers. They have identified and executed on high-value data center sites, often in competitive
constrained power markets. Their internal capabilities span site selection, power procurement, permitting, and scalable design. These
are complemented by deep industry relationships that provide early access to off-market or strategic opportunities. The Company’s
current pipeline reflects this advantage, with multiple active and vetted development prospects. This approach provides a significant
time-to-market advantage over traditional greenfield projects, since it eliminates much of the time and uncertainty associated with
securing new power agreements and vertical construction permits. As a result, the average build time for retrofits, based on senior management’s
experience at Enovum even prior to its acquisition by the Company, is approximately six months from commencement of construction.
We believe, based on our knowledge of the industry, this is approximately one-third to one-half of the industry average development
timeline for greenfield projects.

9

As a result of these and other strategies, we believe we achieve significant
cost advantages. Management estimates that our average build-out cost per MW (gross) is approximately $8 to $10 million, as
compared to an industry average of approximately $13 million per MW (gross).

Modern, strategically located data center
portfolio.

Our MTL-1 and MTL-3
sites, as well as our sites under development, are intended to be modern, high quality facilities built on Tier-3 infrastructure,
with industry leading certifications, including SOC 2 Type 2, reflecting the highest standard of security controls. Our MTL-2, MTL-3 and
NC-1 projects are intended to feature direct-to-chip liquid cooling. We believe this will position us to capitalize on surging
demand for AI and other HPC workloads. Our MTL-1, MTL-2, and MTL-3 sites are located in Montreal, a key market for data centers attributable
to its cold climate, affordable green power, and robust fiber network infrastructure. These factors, collectively, contribute to customer
demand for Montreal facilities. Our NC-1 site, is located in the key data center corridor running along the U.S. East coast,
where eight hyperscaler data centers are located within a 100 mile radius of our facility.

Complementary strategic integration of data
center infrastructure and cloud services businesses.

Our business model includes
integration of HPC data center infrastructure and cloud services to provide scalable, high-performance computing solutions for enterprises,
research institutions, and AI and ML driven businesses. Whether by optimizing cooling strategies for dense GPU clusters or ensuring multi-ISP connectivity
for uninterrupted data flows, we believe our combined HPC data center and cloud services businesses set a higher benchmark for speed,
reliability, and overall client satisfaction in the AI and HPC space. Through vertical integration, we are able to provide power to support
our cloud service needs and also empowered to respond with agility to both emerging hardware trends and the real-time usage metrics
gleaned from our cloud services. Instead of retrofitting third-party data center infrastructure or juggling multiple outside vendors,
we can proactively tailor the entire data center stack to deliver high-performance, future-proof solutions for customers. By integrating
cloud services with our data center infrastructure, we believe we are well-positioned to capitalize on the growing demand for AI
and HPC workloads, capturing margin at both tiers of a vertically integrated value chain and reducing supplier dependency.

Differentiated management team with deep
experience in data center and cloud infrastructure development.

Our management team comprises
individuals drawing on diverse knowledge and skill sets acquired through extensive experience in data centers, real estate and REITs,
and the technology sector. Our leadership also incorporates extensive strategic expertise in finance and capital markets. We augmented
our management team with the acquisition of Enovum with 20 years plus of data center development experience and through other recent
key hires, optimally positioning the team to continue driving our data center and cloud services growth.

Our Growth Strategies

Complete build-out and maximize
revenue from current data center projects.

Leveraging our development capabilities, we intend to complete the
first 27 gross MW (gross) of service at NC-1 in April 2026, with another 27 gross MW (gross) of service in May 2026. These time periods
are based upon the commissioning of the construction and buildout of the facilities. As it relates to MTL-2, we expected to invest approximately
$23.6 million to develop the site to Tier-3 standards with an initial load of 5 MW (gross). However, we have prioritized other builds
and preserved capital for more time sensitive projects.

10

Further, there is potential
to increase revenue from certain of our existing sites by securing additional allocations of utility power, subject to our receipt of
funding and required permits through ongoing engagement with the utility and relevant authorities. We aim to demonstrate the economic
value and readiness of our sites to support incremental load growth, while also monitoring infrastructure upgrades and interconnection
queue developments that may enable expanded allocations over time. Finally, at certain new and existing sites, we intend to deploy natural
gas fuel cell generation technology to increase available power and revenue potential. We expect to achieve this by partnering with third-party energy
providers to deploy natural gas-powered fuel cell systems on or near our data center sites. These systems can be deployed modularly
and connected behind the meter, allowing us to supplement utility power, support additional compute deployments, and increase site revenue
without relying solely on the utility grid. We are currently evaluating proposals and permitting pathways to enable these installations
where economically and operationally viable.

Rapidly and strategically scale our proprietary
data center expansion pipeline.

To complement our existing
HPC data center portfolio, we intend to rapidly develop additional sites from our expansion pipeline. In doing so, we will target selected
locations to secure a strategic presence across North America. We are committed to offering extensive geographical coverage to seamlessly
support our clients’ operations while minimizing latency. By developing a robust HPC data center platform across North America,
we expect to enhance redundancy, mitigate geo-location risks, and ensure our services are available where clients need them most.
We expect our strategically placed WhiteFiber data centers in smaller urban areas, which we term edge data centers, will deliver carrier
hotel-level connectivity, while our larger deployments will power AI-driven computing super-clusters, driving innovation and
efficiency.

Focus on next-generation data
center designs and infrastructure.

Our data center designs feature
high density racks and direct-to-chip cooling architecture to enable our customers’ AI, ML and other HPC workloads. We will
continuously analyze emerging trends in an effort to develop future-proof designs that accommodate increasing densities, while addressing
our clients’ current needs with practical solutions. Leveraging the latest advances in direct-to-chip cooling, we expect to
support both rapidly growing standard clusters and customized high-performance deployments, ensuring efficiency, scalability, and
innovation to meet current and future demand.

Leverage our unique technology strategy
and strategic relationships to grow revenue from existing and new customers.

Our technology team has developed
in-house software to streamline the delivery of cloud services for a wider variety of customers, including those with the highest
performance, reliability, and security requirements. This will allow us to differentiate our cloud services offering from those of other
cloud providers that only offer basic services, such as limited storage, networking and computing performance tiers, and limited delivery
options (e.g., bare-metal, virtualized, or containerized exclusively).

We also intend to leverage
existing strategic relationships to secure new customers and increase revenue from existing customers. We are currently targeting small
and medium-sized customers with high returns. Such relationships include OEM and ODM suppliers and data center planning partners.
For example, we have a co-marketing agreement in place with NVIDIA as a function of being a NVIDIA Preferred Partner through the
NPN and with Hewlett Packard Enterprise, as a function of being accepted into their partnership program. With Super Micro Computer, Inc.,
we have early access to next generation computing platforms for testing and evaluation. We are working with vendors, such as Canopy Wave,
Semper Victus IT, and Trainy.AI, to rapidly plan and deploy innovative HPC cluster designs. Our partnerships enable lead and customer
sharing to rapidly identify customers who will uniquely benefit from our offerings.

11

Prudently source and allocate growth capital.

We have limited leverage,
as we have primarily financed our growth to date with equity. This has provided us with the opportunity to incur prudent leverage to fund
our growth and enhance returns to our shareholders. This was confirmed with our entry into a CAD $60 million (approximately USD $43.8 million)
credit facility in June 2025 with RBC (the “Credit Facility”) as well as the private offering of $230.0 million in aggregate
principal amount of Notes in January 2026, each discussed in more detail below. In our data center business, we believe we are in a position
to utilize well-priced first mortgage and corporate facilities from both Canadian and U.S. commercial banks. Within our cloud
service business, we anticipate utilizing equipment leasing facilities to limit our equity capital commitments. While we will benefit
from access to public equity, we will also continue to explore private equity financings in the form of joint ventures with institutional
partners. We believe that our access to these varied alternatives will provide us the ability to optimize our cost of capital. See “Management’s
Discussion and Analysis of Financial Condition and Results of Operations — Liquidity and Capital Resources.”

Strategic Relationships

Financing

Based on management’s
knowledge of the industry, we believe that there is market demand from institutional private equity investors for exposure to the types
of projects in our data center pipeline and our capability to develop them. We believe that we are well positioned to capitalize on this
demand by forming one or more equity joint ventures. Doing so would provide us access to a differentiated and non-dilutive source
of private equity capital to fund our projects, and deliver a durable stream of cash flows in the form of management fee income for our
services.

In June 2025 we entered
into the Credit Facility with RBC, which provides for up to a CAD $60 million (approximately USD $43.8 million, based on the
CAD/U.S. $ rate of exchange of CAD 1.00/U.S. $0.7308, as reported by Bloomberg on June 18, 2025) RBC and the Company have
since entered into discussions to amend the Credit Facility to utilize the proceeds CAD $24.5 million (approximately USD $17.9 million)
from the mortgage loan to purchase the MTL-3 facility, however, we cannot assure you that the parties will reach an agreement that is
acceptable to the Company or at all. See “Management’s Discussion and Analysis of Financial Condition and Results of Operations
– Overview” for more information.

On January 26, 2026, we completed a private offering of $230.0 million
aggregate principal amount of 4.500% Convertible Senior Notes due 2031 (the “Notes”), including the exercise in full
of the initial purchasers’ option to purchase an additional $20.0 million aggregate principal amount of Notes. We used approximately
$120.0 million of the net proceeds from the Notes offering to pay the cost of the zero-strike call option transaction we entered into
simultaneously with such option, and the remaining net proceeds of the offering primarily for data center expansion, including to partially
fund the lease or purchase of additional property or properties on which to build additional data centers, to construct those facilities,
to enter into additional energy service agreements for each additional site, to purchase related equipment, and for potential acquisitions,
partnerships and joint ventures related thereto, and for working capital and general corporate purposes. We will require additional project
financing (e.g., construction loans) in order to fully accomplish such initiatives. We also may elect to raise additional capital opportunistically.

On March 25, 2026, WhiteFiber Iceland ehf. (the “Borrower”),
a subsidiary of the Company, entered into a secured term loan facility agreement (the “Facility”) with Landsbankinn hf, which
provides for borrowings of up to $20 million. Borrowings under the Facility bear interest at a floating rate per annum equal to the sum
of (i) three month CME Term SOFR (or any successor benchmark), and (ii) an applicable margin of 4.25% per annum. The Facility is secured
by first-ranking security over (i) 100% of the Company’s shareholding in WhiteFiber Iceland ehf., (ii) designated assets (including
GPU servers, CPU servers, IB switches and equipment accessories) at the date of the agreement, and (iii) material assets acquired thereafter
(to be secured within 60 days), in each case until all obligations are fully satisfied.

12

See “Management’s
Discussion and Analysis of Financial Condition and Results of Operations — Liquidity and Capital Resources”.

Technology

We have established formal
relationships with leading technology providers, including NVIDIA, Super Micro, Dell, Hewlett Packard Enterprise, Super Micro and Quanta
Computing. These partnerships enable us to access and deploy the most advanced computing hardware, ensuring that our clients benefit from
cutting-edge technology and optimized performance. Our collaboration with these industry leaders allows us to stay at the forefront
of high-performance computing and AI infrastructure.

In addition, we have a formal
agreement with Shadeform to offer on-demand compute services powered by NVIDIA B200 GPUs. This partnership enhances our ability
to provide scalable, high-performance AI and HPC solutions to enterprise customers, enabling them to access state-of-the-art computing
resources without the need for upfront infrastructure investment.

Our data center business
leverages a diverse mix of vendors across multiple geographies to source equipment cost-effectively while mitigating supply
chain disruptions. By maintaining a broad supplier network, we reduce reliance on any single vendor and enhance our ability to procure
best-in-class solutions at competitive prices. This approach strengthens our operational resilience and supports our ability to scale
efficiently.

Competition

We face significant competition
from various data center and cloud services providers. We compete with several prominent data center providers, including Digital Realty,
Equinix, Inc., NTT, Cyrus One, Inc., STACK Infrastructure, Inc., Aligned Data Centers, LLC, Iron Mountain and various private operators
in the U.S. Our primary competitors in the cloud service business are CoreWeave, Crusoe Energy, Nebius, and Lambda Labs.

Many of our competitors offer
locations worldwide and have well-established international operations. Our competitors may also have significant advantages over
us, including greater name recognition, longer operating histories, pre-existing relationships with global developers, utilities
and local authorities, which may give them an advantage in securing real estate or power for new sites, particularly in constrained regions,
the capacity to provide the same or additional products and services, more significant marketing budgets and other financial and operational
resources, more robust internal controls and systems, and better established, more extensive scale and lower cost suppliers sand supplier
relationships. In addition, as we develop proprietary technology and software solutions to support AI and HPC workloads, we face competitive
risks from larger incumbents with greater R&D resources and established ecosystems. The competitors may bring similar offerings to
market faster or with deeper integration into existing platforms, potentially limiting our market share or pricing power.

Materials and Suppliers

Maintaining key supplier
relationships is crucial to our business operations, as we rely on these relationships, such as with Dell, NVIDIA, Hewlett Packard Enterprise,
Super Micro and Quanta Computing, to secure GPUs, servers, essential computing hardware, infrastructure components, and other materials.
The complexity of developing cloud service hardware at scale limits the number of suppliers capable of meeting our requirements. The development
of new data center capacity is subject to supply chain constraints and long lead times for critical infrastructure components such as
power distribution equipment, generators, and cooling systems. We work with experienced vendors and maintain forward-looking procurement
plans to mitigate these risks, but delays could still impact project timelines. Consequently, we have established purchase orders with
leading hardware manufacturers that include extended delivery schedules spanning several months before the hardware is delivered
to our facilities. These fluctuations in delivery timelines necessitate careful planning and advanced purchasing strategies to ensure
we can acquire hardware well before their anticipated deployment.

13

We proactively procure these
materials from our suppliers in sufficient quantities to facilitate hardware deployment at scale and on accelerated timelines. To mitigate
potential supply chain disruptions and ensure the smooth operation of our facilities, we have established long-term contracts and
agreements with key suppliers. This includes multi-quarter purchase commitments for critical hardware such as GPUs, power distribution
units, and networking equipment, as well as ongoing service agreements with construction, electrical, and facility operations vendors.
These relationships help secure allocation priority, stabilize pricing, and reduce lead-time risk for both cloud infrastructure and
data center development. These arrangements give us greater certainty regarding the availability and pricing of essential components and
materials. Furthermore, we continuously monitor market trends and maintain open lines of communication with our suppliers to anticipate
and address potential supply chain challenges.

By proactively managing our
supplier relationships, securing necessary materials in advance, and closely monitoring market conditions, we aim to minimize the impact
of supply chain fluctuations on our operations. This approach enables us to maintain a steady pace of hardware deployments and facility
development, ultimately supporting our goal of expanding our HPC data center and cloud service capabilities and maximizing shareholder
value. However, we rely on a limited number of vendors for certain products and services for our data center facilities, and some of our
contracts provide a single source of materials. If any of our key suppliers cannot perform under their contracts or satisfy our orders,
it could significantly delay our data center development and operations. While we may be able to engage replacement suppliers, this would
likely lead to operational delays and increased costs.

Power Supply

In the province of Quebec where MTL-1, MTL-2, and MTL-3 are located,
all of the hydroelectric power is provided by a crown corporation, Hydro Quebec, which has predetermined rates depending on the customers’
industry and based on the power demand.

As set forth above, the Company
has entered into a Capacity Agreement with Duke Energy for the Company to receive 24 MW of service to NC-1 by September 1, 2025,
an additional 40 MW by April 1, 2026 and final phase of permanent service 99 MW within four years of May 16, 2025. The actual rates
will be determined when the facilities are turned on.

Customers

We generate a large portion
of our revenue from a small number of customers. There are inherent risks whenever a large percentage of total revenue is concentrated
with a limited number of customers. If we were to lose one or more of our customers, our operating results could be materially adversely
affected. For more information, see “Risk Factors – Our business has and is expected to continue to have significant customer
concentration.”

14

WhiteFiber data centers

Our HPC data center customer
base consists of two primary types of customers:

●Enterprise clients — current
and prospective enterprise clients are active in multiple industries, including healthcare, finance, and various technologies that rely
on computers or models. These customers benefit from our high-density solutions, reaching upwards of 50kW per cabinet, to accommodate
their workloads and data generation.

●GPU cloud — Our GPU cloud customers
offer on demand access to their GPUs for tasks like AI, VFX rendering and scientific computing.

Currently, we provide HPC
data center services at our leased MTL-1 and MTL-3 facilities, although our customers are based across Canada and Europe. As of December
31, 2025, our leased MTL-1 and MTL-3 facilities served fifteen customers. No one customer accounted for in excess of 50% of data
center revenue in the 12 months ended December 31, 2025 or 2024.

Cloud Services

Our cloud services customer
base also is comprised of two primary types of customers:

●Direct end users — these customers
primarily leverage our computing power for model training and inference.

●GPU marketplaces — these platforms
resell our computing power to their own end users. Since we do not have direct visibility into their end-user base, there may be
some overlap among end users across different marketplaces.

We currently provide cloud services at Blönduos Campus, Iceland,
where we lease capacity to house our GPUs. During 2025, our HPC data center in Iceland had contracts with twenty one customers. DNA Fund
accounted for approximately 11.5% of our revenue during the 12 months ended December 31, 2025. We did not generate any revenue
from DNA Fund in 2024. Our Initial Customer accounted for approximately 70.7% of our revenue during the 12 months ended December 31,
2025 and 96.6% of our revenues through December 31, 2024. As of the report date, the Company and the Initial Customer are engaged
in ongoing discussions regarding a potential resolution of the existing service agreements following the agreed pause of services. No
definitive agreement has been reached. Following the service pause, the Company has redeployed the GPUs previously allocated to the Initial
Customer to three other customers. Based on our current customer mix and contracted capacity for 2026, a limited number of customers are
expected to represent a significant portion of our cloud services revenue. While we are actively expanding and diversifying our customer
base, our results may continue to be influenced by the performance and contractual arrangements of these customers.

We have leased additional
capacity to install our data center in Atlanta, Georgia, USA to expand our cloud services offering. The capacity leases commenced in February
2026. We also intend to lease additional capacity to expand our cloud services offering.

Global Logistics and Tariffs

Global supply logistics have
caused delays across all distribution channels, impacting the HPC, AI and ML markets. Delivery schedules for specialized equipment, such
as high-performance computing systems, AI hardware, and necessary infrastructure components, have been affected due to constraints
on globalized supply chains. These constraints extend to procuring construction materials and specialized electricity distribution equipment
required to develop HPC, AI and ML facilities. Efforts to mitigate delivery delays are ongoing to avoid materially impacting deployment
schedules; however, there are no assurances that such mitigation efforts will continue to be successful. To help address global supply
logistics and pricing concerns, we have implemented proactive measures such as procuring and holding required materials. We continuously
monitor developments in the global supply chain which is necessary to assess their potential impact on our expansion plans within the
HPC, AI and ML markets. This monitoring includes evaluating the impact of international trade policies and tariffs, which may affect the
cost or availability of key components source from abroad and, in turn, impact our expansion timelines or capital expenditure.

15

Data center construction
relies heavily on steel, copper, aluminum, electrical components and HVAC systems, some of which the Company is sourcing from Mexico and
Canada. The tariffs the U.S. has imposed, or has considered imposing, on Canadian steel, aluminum and copper imports, are expected
to increase the cost of WhiteFiber’s potential projects in the U.S. Similarly, reciprocal tariffs imposed by Canada on WhiteFiber’s
projects in Canada on U.S. exports could see cost increases for imported power infrastructure, networking hardware, and construction
equipment. See “Risk Factors – Changes in tariffs or import restrictions could have a material adverse effect
on our business, financial condition and results of operations.”

Regulatory Landscape

The regulatory landscape
surrounding WhiteFiber data centers and cloud services is evolving rapidly, and we anticipate increased scrutiny and potential regulation
in the near- and long-term. These developments may have a material adverse effect on our business and financial condition.

We are subject to the laws
and regulations of various jurisdictions and governmental agencies affecting our operations and the sale of our infrastructure and services
in areas including, but not limited to: AI, intellectual property; tax; import and export requirements; anti-corruption; economic and
trade sanctions; national security and foreign investment; foreign exchange controls and cash repatriation restrictions; data privacy
and security requirements; competition; advertising; employment; product regulations; environment, health and safety requirements; and
consumer laws.

Although there is no assurance
that existing or future governmental laws and regulations applicable to our operations and the sale of our infrastructure services will
not have a material adverse effect on our capital expenditures operating results and competitive position, we do not currently anticipate
material expenditures for complying with government regulations. Nevertheless, we believe that global trade regulations could potentially
have a material impact on our business.

As a global company, the
import and export of our infrastructure services and technology are subject to laws and regulations including international treaties,
U.S. export controls and sanctions laws, customs regulations, and local trade rules around the world. For example, all acquisitions
of control (whether direct or indirect) of Canadian businesses by non-Canadians may be subject to review on grounds that the investment
could be injurious to national security. Following the filing of the required Investment Canada Act notification in respect of our acquisition
of Enovum, we received written notification from the Minister of Innovation, Science and Industry of Canada (the “Minister”),
that the acquisition may be subject to a national security review. Following review of information we provided, Bit Digital executed a
“Commitment Letter” to the Minister which provides for the following: (i) commitment to maintain one Canadian on the
board of Enovum Inc.; (ii) commitment to maintain or improve Enovum’s security controls for personnel, physical security and
the internal network, and (iii) commitment to send a list of Enovum’s current clients to Investment Canada on an annual basis.
The Company is complying with the terms of the Commitment Letter.

The scope, nature, and severity
of such controls varies widely across different countries and may change frequently over time. Such laws, rules, and regulations may delay
the introduction of our infrastructure and services or impact our competitiveness through restricting our ability to do business in certain
places or with certain entities and individuals. For example, the U.S. Department of Commerce continues to add firms to the Entity
List. These export restrictions, which would require that we obtain licenses from the U.S. Department of Commerce to allow us to
export infrastructure services to such listed firms, which could limit or prevent us from doing business with certain potential customers
or potential suppliers. Additionally, although the U.S. Department of Commerce has withdrawn its AI diffusion rules, which would
have imposed worldwide limits on AI chip exports used to create computer clusters, it plans to issue a replacement rule in the future
and continues to strengthen other existing export controls on advanced AI chips. These restrictive governmental actions, and any similar
measures that may be imposed on U.S. companies by other governments, could limit our ability to conduct business globally.

Additionally, there are growing
concerns about the ethical implications and potential misuse of AI and machine learning. Governments and regulatory bodies are considering
measures to ensure the responsible development and deployment of AI systems, including transparency, accountability, and fairness guidelines.
We are closely monitoring these developments and will dedicate our best efforts to adhere to any upcoming regulations or industry best
practices.

As a company operating at
the intersection of data center, cloud and HPC hosting services, we are committed to maintaining a proactive and adaptive approach to
regulatory compliance. We closely monitor legislative and regulatory developments and engage in dialogue with relevant stakeholders to
ensure our business practices align with the evolving legal and regulatory framework. Despite the uncertainties posed by the changing
regulatory landscape, we remain committed to delivering innovative and responsible solutions in the data center, cloud and HPC hosting
markets while prioritizing compliance and risk management. However, if we fail to comply with applicable laws and regulations, we may
be subject to significant liabilities, including fines and penalties, and our business, financial condition, or results of operations
could be adversely affected.

16

Investment Canada Act

All acquisitions of control
(whether direct or indirect) of Canadian businesses by non-Canadians may be subject to review on grounds that the investment could
be injurious to national security. Following the filing of the required Investment Canada Act notification in respect of our acquisition
of Enovum (the “Transaction”), we received written notification from the Minister of Innovation, Science and Industry of Canada
(the “Minister”), that the Transaction may be subject to a national security review. Following review of information we provided,
Bit Digital executed a “Commitment Letter” to the Minister which provides for the following: (i) Commitment to maintain
one Canadian on the board of Enovum Inc.; (ii) Commitment to maintain or improve Enovum’s security controls for personnel,
physical security and the internal network, and (iii) Commitment to send a list of Enovum’s current clients to office of the
Minister of Innovation, Science and Industry on an annual basis. The Company is complying with the terms of the Commitment Letter.

Employees

As of December 31, 2025, we and our subsidiaries employed 83 full-time employees
who support our cloud and data center operations, infrastructure development, and corporate functions. We focus on attracting and retaining
skilled technical and operational personnel through competitive compensation, benefits, and performance-based incentives aligned with
our business objectives. Management oversees human capital matters, with an emphasis on workforce safety, operational reliability, and
compliance with applicable labor and employment laws. We also engage consultants and other third-party providers to support our business
activities.

In connection with the completion of our initial public offering, we
implemented a long-term performance incentive program, granting eligible employees service-based restricted stock awards and
performance-based restricted stock awards that vest upon achieving specific performance milestones. This performance program is a
key employee incentive, aligning their long-term interests with the Company’s objectives. See Item 11 - Executive Compensation
and Director Compensation .

Corporate Information

We were incorporated by Bit
Digital as a Cayman Islands exempted company on August 15, 2024 under the name Celer, Inc., as a holding company for the HPC Business.
We changed our name to WhiteFiber, Inc. on October 17, 2024. We completed our IPO on August 8, 2025, and our Ordinary Shares are currently
listed on Nasdaq under the symbol “WYFI”.

Our executive office is located
at 31 Hudson Yards, Floor 11, Suite 30, New York, New York 10001, and our phone number is (646) 801-0779.

Available Information

Our Annual Reports on Form
10-K, Quarterly Reports on Form 10-Q, Current Reports on Form 8-K, and amendments to reports filed pursuant to Sections 13(a) and 15(d)
of the Exchange Act, are filed with the U.S. Securities and Exchange Commission (the “SEC”). We are subject to the informational
requirements of the Exchange Act and file or furnish reports, proxy statements, and other information with the SEC. Such reports and other
information filed by us with the SEC are available free of charge on our website at www.whitefiber.com when such reports are available
on the SEC’s website at www.sec.gov.

We announce material information
to the public through filings with the SEC, the investor relations page on our website at www.whitefiber.com, press releases, public conference
calls and public webcasts. The information disclosed through the foregoing channels are intended to be sources of material information
about the Company. As such, we encourage investors, the media and others to follow the channels listed above and to review the information
disclosed through such channels. Any updates to the list of disclosure channels through which we will announce information will be posted
on the investor relations page on our website.

The information contained
on, or accessible through, our website is not a part of, and is not incorporated into, this Annual Report. We have included our website
address only as an inactive textual reference and do not intend it to be an active link to our website. You should not rely on our website
or any such information in making your decision whether to purchase our Ordinary Shares.

17