4DS Memory Jenesys logo JENESYS
4DS Memory Limited (ASX:4DS) · Confidential — September 2026
Building Next-Gen
Edge AI Capability

Investor Presentation|4DS Memory Limited’s proposed A$5 million capital raise to fund the acquisition of Jenesys and accelerate its expansion into Edge AI and autonomous defence technologies.

← → Navigate
Jenesys logoJENESYS
Autonomous Systems for Defence
Edge AI enabling multiple platforms across every domain to act as one coordinated force.

One device that thinks is useful. A fleet that thinks together is asymmetric advantage.

We build both halves: the intelligence that runs on each platform, and the layer that makes them a team.

Jaspal Sarai·Jenesys·jenesys.com.au
← → Navigate

Important Notice & Disclaimer.

Purpose & Offer

This presentation has been prepared by 4DS Memory Limited (ASX:4DS) in connection with a proposed capital raising of approximately A$5 million (Capital Raising) to fund the proposed acquisition of Jenesys and provide working capital to advance the combined business.

The Capital Raising will be conducted in accordance with applicable requirements of the Corporations Act 2001 (Cth). This presentation does not constitute a prospectus or other disclosure document and has not been lodged with ASIC. Investors should not subscribe for securities except on the basis of the terms and conditions of the Capital Raising and should make their own assessment of 4DS.

Information & Investment Risk

The information in this presentation is provided in summary form and does not contain all information necessary to make an investment decision. It does not constitute financial, legal, taxation or investment advice and does not take into account any investor's objectives, financial situation or needs. Investors should conduct their own enquiries and obtain appropriate professional advice before making an investment decision.

An investment in 4DS is subject to risks, including those associated with the proposed Jenesys acquisition, technology development and commercialisation, funding requirements and market conditions. No representation or warranty is given as to the accuracy, completeness or future performance of the Company, except to the extent required by law.

Forward-Looking Statements

This presentation contains forward-looking statements regarding the proposed acquisition, technology development, commercialisation, market opportunities and future business plans. These statements are based on current expectations and assumptions and are subject to risks and uncertainties. Actual outcomes may differ materially and no guarantee is given that any forward-looking statement will be achieved.

Illustrative Material

Certain diagrams, images and application scenarios are illustrative only and should not be interpreted as representing existing commercial deployments unless expressly stated.

Distribution

This presentation is confidential and may only be distributed to persons entitled to receive it in connection with the Capital Raising. It must not be copied, reproduced or distributed without the prior written consent of 4DS.

Capability status labels

Capabilities in this presentation are marked:

Operational
In service.
Demonstrated
Built and shown to work in its intended environment.
Prototype
Built and bench-validated, not yet proven in its operational environment.
In Development
On the current roadmap.
Target Application
An intended use case, not a delivered capability.

Technology Readiness Levels are stated per asset, not for the Company as a whole.

Advanced Memory for the Emerging Edge AI Ecosystem

4DS Memory

Advanced Interface Switching ReRAM architecture designed for future memory and compute environments — positioned within the evolving landscape of advanced memory and Edge AI infrastructure.

34
Patents

Interface Switching ReRAM IP portfolio

A$60M+
R&D Investment

Over a decade of semiconductor development

60nm
Demonstrated Cells

CMOS-compatible process pathway

Tunable
Endurance/Retention

Advanced switching architecture characteristics

Building device stack
Technology Position

4DS Memory has developed a proprietary ReRAM technology platform demonstrated at the 60nm cells — a process level aligned with the mature-node semiconductor manufacturing capabilities being established across emerging fabrication markets.

ProcessCMOS-compatible architecture suitable for mature-node fabrication pathways
ApplicationsIndustrial, edge AI, aerospace, defence and embedded systems
Areas of Strategic Focus

Advanced memory architectures are increasingly relevant to emerging Edge AI infrastructure requirements.

Persistent Memory

Next-generation persistent memory architectures.

Compute Efficiency

Low-power memory approaches for future AI systems.

Edge AI Relevance

Exploring applicability to distributed AI infrastructure.

Semiconductor Scalability

Long-term architecture and integration pathways.

What is Edge AI? Edge AI brings artificial intelligence directly to devices — allowing them to process information, make decisions and respond in real time without relying on a constant connection to the cloud.

Edge AI Software for Autonomous Defence Systems

Jenesys

Jenesys develops software that enables drones, vehicles and other autonomous systems to make decisions locally and work together — without relying on constant connectivity or an individual operator for every platform.

About Jenesys

Sovereign Australian Capability

Australian defence technology company developing sovereign Edge AI and autonomous systems software.

Two Technology Lines

  • Edge AI: enables AI models to run directly on drones, vehicles and other platforms.
  • Autonomous Coordination: enables multiple platforms to coordinate and divide tasks between themselves.

Software Only

Designed to run on existing drones, vehicles and other systems — without Jenesys needing to manufacture the underlying hardware.

Separated IP Tracks

Defence and commercial technologies are maintained on separate development pathways, protecting sovereign defence capability while enabling broader commercial applications.

Command the mission, not each platform.

Jenesys puts intelligence where data is captured.

Platform-agnostic by design. Our edge AI runs on hardware already in the field, and the coordination layer accounts for their invididual capabilities across air, land, sea and space. A suite of software IPs allow us to bring this heterogenous autonomous swarming capability to life.

Model optimisation pipeline · TRL 9 Payload control software · TRL 6–7 Swarm coordination · TRL 3–4 · simulation Onboard data processing · TRL 3–4
Air
Quadcopter UAV
Fighter Aircraft
Land
Quadruped
Armoured Tank
Sea
Surface Vessel
Submarine
Space
Satellite
Initialising

Data is captured at the edge, yet the actionable decisions are made on the ground.

Sensors in the field capture far more information than their connection bandwidth can carry in one transmission. Today's default is to transmit raw data back — to a ground station, a command and data centre — and wait.

L0 raw capture — noisy
L0 Raw
→
L1 denoised
L1 Denoised
→
L2 corrected
L2 Corrected
→
L3 analysed
L3 Analysed
→
L4 insight mask
L4 Insight

The sensors capture gigabytes worth of data, yet the actionable insight from it is worth kilobytes. An entire correction-and-analysis pipeline, paired with computer vision AI models is required to turn these gigabytes of raw capture data into kilobytes of insight — and currently, all of it ships to the ground first.

Latency

Hours from capture to insight — downlink windows, processing queues, analysis backlogs — while the situation on the ground moves on.

Bandwidth

Raw sensor data outruns the link. A single satellite pass captures gigabytes; most of it never reaches the ground at all.

Connectivity

No link, no intelligence. The places that need answers most — oceans, disaster zones, remote country — are exactly where connections fail.

Sovereignty

Your data lands on someone else's infrastructure and is processed in someone else's jurisdiction.

Move the intelligence to the source of data — where time-critical operations matter most.
Model optimisation pipeline · TRL 9

A broad range of AI models, optimised to run on constrained edge hardware.

Space-Ready AI Model Optimizer — one full-size network distilled into small task-specific models over time
Unmanned Autonomous Platforms
Initialising

A proprietary optimisation pipeline takes AI computer-vision models from the standard frameworks, prunes and quantises them, and compiles them into a lightweight C++ runtime — small enough for the constrained, rugged hardware where the data actually lives.

One full-size network becomes a family of small, task-specific models — each fast enough to run on the device itself.

TensorFlow · PyTorch · ONNX 5–10× typical (up to 166×)1 NVIDIA Jetson · NVIDIA GPU · CPU

1 Typical 5–10× end-to-end through the optimisation pipeline (automated pruning and quantisation). The 166× best case is a binary smoke-classification model (whole-image, not per-pixel): 10.0 s → 0.06 s single-image inference, Python/TensorFlow versus the same model put through the proprietary optimisation pipeline and executed in the Jenesys C++ runtime, on identical hardware (NVIDIA GTX 1080) and the same input image — a runtime and format change, not a change to the network, so classification outputs are unchanged. Accuracy after pruning and quantisation is model-dependent and is validated per deployment. Internal testing; not independently verified. Results vary by model, hardware and workload.

Payload control software · TRL 6–7

Deployed, managed and improving — in the field.

Continuous ML On-Board — The Learning Loop
Continuous ML on-board: input image through the model to a fire prediction, validated by people, then used to retrain the model

The learning loop: models run inference on-board; operators validate the outputs; the model is retrained on-board from the outputs they confirm. Human validation gates every update — the system only learns from what is confirmed.

Runs a fleet of models Prototype

Multiple AI models run concurrently on one device — real-time inference, health monitoring and telemetry included.

Secure updates Prototype

New models uplink to hardware that cannot be physically reached — encrypted, with automatic rollback if an update fails.

Humans in the loop Operational

Operators validate the outputs; the model is only updated from what they confirm.

Learns on the platform Prototype

Retraining runs on-board from operator-validated outputs — raw sensor data never has to leave the device.

Payload control software · TRL 6–7

Nexus-01 — Edge AI, deployed to orbit.

Jenesys has already delivered Edge AI hardware and software to orbit, launching its Nexus-01 payload aboard a SpaceX Falcon 9 mission in August 2025 — demonstrating the ability to take AI technology from development through integration, flight qualification and deployment.

Render of the LEAP-01 small satellite carrying the Nexus-01 payload
1 · Payload built Complete
2 · Bus integration Complete
3 · Flight acceptance Complete
4 · Launch manifest Complete
5 · Delivered to orbit Complete

From concept to flight article — through integration with a third-party satellite bus, flight acceptance, a Falcon 9 rocket and into orbit. The flight software and the C++ edge-runtime discipline behind it carry forward into everything we deploy.

L0 raw capture
L0 Raw
→
L1
L1 Denoised
→
L2 corrected
L2 Corrected
→
L3 analysed
L3 Analysed
→
L4 insight mask
L4 Insight

Raw capture to actionable insight in one pass — geometric, radiometric and atmospheric corrections, then AI analysis, all designed to run on the satellite itself.

One pipeline · Many missions — model outputs on public satellite imagery
Smoke plume→Fire mask
Fire Detection
Bushfire early warning
Airport apron→Anomaly heatmap
Anomaly Detection
Force protection & infrastructure watch
Aerial scene→Segmentation map
Segmentation
Mapping & urban analysis
False-colour fields→Classified fields
Crop Classification
Agriculture
Facility→Spectral image
Spectral Analysis
Nuclear facilities & military infrastructure
Initialising
Illustrative — conceptual visualisation of platforms flying fixed, pre-programmed routes without a coordination layer. Not a recording of a physical system.
Proof Points

Proven Foundations, One Integrated Stack

4DS Memory Jenesys

4DS and Jenesys bring together complementary memory and software technologies to create an integrated Edge AI platform — from the underlying memory and compute architecture through to AI and autonomous coordination software.

Building integrated stack
4DS ReRAM
Semiconductor Memory
Jenesys
Coordination Software
Current Status
Proven ReRAM platform demonstrated at the 60nm cells
Algorithmic coordination platform validated in a simulation environment (TRL 4)
Core Capability
Persistent semiconductor memory architecture
Distributed decision-making, task allocation and autonomous coordination software
Key Outcome
Successful megabit array demonstration and patented memory technology platform
Successful autonomous task identification, allocation and mission execution logic validation
Commercial Relevance
Semiconductor memory capability
Software and systems capability

The proposed integration brings together complementary hardware and software assets to support future commercial deployment opportunities.

Initialising
Illustrative — conceptual visualisation of the edge-AI module deploying onto a platform. Not a recording of a physical system.
Swarm coordination · TRL 3–4 · Simulation

Edge AI detects. Swarm collective actions.

Hover over each behaviourTap or scroll to explore

Jenesys is developing autonomous coordination software designed to enable multiple platforms to assess changing conditions, allocate tasks and respond together — without requiring a central operator to direct every action.

Illustrative target behaviours on the Jenesys development roadmap (Q4 2026 – Q1 2028). Currently TRL 3–4 and validated in simulation.

Initialising live demo...

One Coordination Platform.
Many Mission Applications.

Jenesys’ Edge AI and autonomous coordination software is designed to work across multiple missions and operating environments — from disaster response and border surveillance to complex defence applications.

Counter-UASTarget Application
Detect, classify and track hostile drones on-board — and coordinate the response as a fleet, not one operator per threat.
Signal & Spectrum AwarenessPayload-enabledTarget Application
Detect, classify and geolocate transmitters across a wide area — a coordinated grid of RF payloads cross-fixing emitters a lone receiver misses, from covert terminals to hostile radars.
Persistent ISRTarget Application
Persistent, wide-area watch — many low-cost nodes flagging the movement and change a single platform can't sustain.
Border & Maritime AwarenessTarget Application
Flag incursions across coastlines, straits and archipelagos — coordinated across multiple platforms automatically.
Contested-Comms AutonomyIn Development · Q1 2028
Designed to keep operating when GPS and comms are degraded — decisions stay on the platform, with no central node to lose.
Disaster & Environmental ResponseEdge detection demonstrated
Fire, flood and search-and-rescue — the on-board edge detection built for Nexus-01, extended across a coordinated fleet.

Jenesys is the intelligence and coordination layer — it distributes platforms, coordinates tasks best suited to the capabilities of each platform. The applications above are target use cases enabled by partner payloads; labels reflect the maturity of the underlying Jenesys capability.

Initialising
Illustrative mission scenario — a modelled concept of operations, not a flown mission. RF sensing is a partner payload; Jenesys provides the detection-fusion and coordination layer.

Extending Autonomous Coordination Beyond Defence

Jenesys’ technology has potential applications wherever fleets of robots and autonomous systems need to work together — creating opportunities across multiple large and growing commercial markets.

Building Management & Cleaning

Cleaning and service robot fleets tasked alongside human cleaning staff.

Professional cleaning-robot sales +34% to >25,000 units, 20241

Warehouse & Intralogistics

Mixed-vendor robot fleets under one coordination layer, not a console per brand.

>50% of companies deploying intralogistics robots on a multiagent orchestration platform by 20262

Infrastructure Monitoring

Power, rail and pipeline networks swept as coordinated coverage, not single sorties.

Over 80m km of grid to add or replace by 2040 to meet national energy goals3

Search & Rescue

Automatic wide-area search — the fleet spreads out based on individual capabilities.

1,000+ people rescued by drones across 39 countries, 20234

Disaster & Emergency Response

Fire, flood and storm assessment that holds together when the network does not.

US$220bn economic losses from natural catastrophes, 20255

Agriculture & Land Management

Mixed machinery across a field, allocating work by capability and position.

~19,500 agricultural robots sold worldwide, 20241

Mining & Resources

Mixed-vendor autonomous fleets coordinating across a single site.

Komatsu commissions its 1,000th ultra-class autonomous haul truck, Apr 20266

Ports & Terminals

Yard vehicles, handling equipment and inspection assets work under one system.

76 container terminals with some form of operational automation, 20257

Law Enforcement & Public Safety

Responding robot assets self-allocating to incidents and reporting quicker to authorities.

1,000+ US agencies cleared to fly BVLOS for drone-first-response in 10 months8

Crowd & Event Safety

Persistent watch over large sites, with coverage re-planned as the crowd moves.

US$2.96bn drone crowd-monitoring market forecast, 20269

Jenesys is the intelligence and coordination layer — it does not build the robots, sensors or platforms it coordinates. The markets above are target applications, not current deployments; each is addressed by applying the same capability-based allocation to fleets built by others. Figures are third-party market and industry data, not Jenesys performance, and are cited in full below.

1 International Federation of Robotics, World Robotics 2025 — Service Robots, 7 Oct 2025; 294-supplier sample, not projected to the whole industry. 2 Gartner, Predicts 2023: Supply Chain Technology, 28 Nov 2022 — a 2022 forecast; Gartner has published no assessment of the outcome. 3 International Energy Agency, Electricity Grids and Secure Energy Transitions, Oct 2023; requirement if announced national energy and climate goals are met, all voltage levels. 4 DJI, DJI Records More Than 1,000 People Rescued By Drones Globally, 11 Jul 2023, and DJI Drone Rescue Map; vendor-compiled from public reports. 5 Swiss Re Institute, sigma No 1/2026, 19 Mar 2026; insured losses US$107bn. 6 Komatsu, Komatsu becomes first OEM to commission 1,000 ultra-class autonomous haul trucks, 21 Apr 2026; cumulative units commissioned since 2008. 7 Port Economics, Management & Policy, Automated Container Terminals, 2025; data from Rodrigue, J-P (2025), Journal of Shipping and Trade 10(1), art. 4. 50 of the 76 are semi-automated only. 8 FAA Part 91 BVLOS waivers, Apr 2025 – Feb 2026; FOIA-released list published by the Electronic Frontier Foundation, 23 Jul 2026. A waiver is the regulatory prerequisite for a DFR program, not a launched program. 9 The Business Research Company, Drone-Assisted Crowd Monitoring Global Market Report 2026; single-source syndicated forecast.

Commercial Model

Accelerating Commercial Execution

Combining semiconductor memory, software capability and market access to support commercial deployment within emerging semiconductor markets.

4DS Foundation
  • Proprietary 34-patent PCMO ReRAM IP portfolio
  • Proven 60nm ReRAM technology platform
  • Established fabrication process expertise
  • Electrical switching and control methodology
+
Jenesys Enablement
  • Software integration and SDK capability
  • AI/ML software and algorithm expertise
  • Edge AI systems architecture capability
  • OEM and channel engagement relationships
  • Market presence and execution capability
=
Commercial Outcome
  • Integrated semiconductor and software platform
  • Positioned for engagement with semiconductor fabs and OEM ecosystems
  • Enhanced customer adoption and deployment capability
  • Capability aligned to edge AI, industrial and strategic technology applications
  • Potential compute-in-memory research and development pathway
Why a combined solution?

4DS provides the semiconductor memory technology, while Jenesys contributes software, systems integration and market engagement capability. Together they create a more complete platform designed to support customer adoption, product integration and commercial deployment.

Why integrate rather than build internally?

4DS believes building equivalent software integration and OEM enablement capability internally would require significant time, investment and execution resources. The integration is intended to accelerate commercial readiness by adding established software capability, systems integration expertise, OEM relationships and market presence.

Strategic Pipeline

Relationships and engagement pathways providing access to priority defence, industrial and semiconductor sectors.

Opportunity / SectorDescriptionStatus
ASCA Emerging & Disruptive Technologies
Defence

Australia's Department of Defence program investing in long-term R&D partnerships with sovereign industry and academia to develop technologies that deliver asymmetric advantage to the ADF. With over $100 million committed across successive rounds, current priority areas include Decision Advantage, Quantum Technologies, Information Warfare, and Persistent and Pervasive ISR (P2ISR).

Early engagement
Global Prime Engagement
Defence

Global aerospace and defence technology leader operating across communications, autonomous systems, space and national security markets. Potential pathway for advanced autonomous coordination and edge-AI applications.

Early discussions
Royal Australian Air Force (Australian Defence Force)
Defence / Government

Australia's air force and a leading adopter of advanced aerospace, autonomous and mission systems technologies. Potential applications for autonomous coordination, edge-AI and resilient communications capability within future defence environments.

Early discussions
Strategic Rationale

The integration of Jenesys is intended to enhance 4DS' commercial deployment capability through software, systems integration and market engagement pathways aligned with the Company's broader industrial, defence and infrastructure opportunities.

Key Leadership & Personnel

Bringing together deep expertise across semiconductors, global technology commercialisation, capital markets and advanced AI — including award-winning leadership in autonomous systems and machine intelligence.

4DS Memory — Board & Management
David McAuliffe
David McAuliffe
Executive Chairman

Corporate executive with more than 24 years' experience founding, building and scaling ASX-listed and international businesses. Extensive expertise in corporate strategy, capital markets, business development and commercial execution. Holds a Bachelor of Laws (Honours) and a Bachelor of Pharmacy.

Howard Digby
Howard Digby
Non-Executive Director

More than 25 years' experience in technology leadership across the Asia-Pacific region. Former executive with IBM, Adobe and Gartner, with extensive experience advising global technology organisations including Samsung, SK Hynix and TSMC. Holds a Bachelor of Engineering (Honours) from The University of Western Australia.

Guido Arnout
Guido Arnout
Non-Executive Director

More than 30 years' experience commercialising advanced technologies, including guiding multiple Silicon Valley companies through growth, strategic partnerships and successful trade sales. Extensive expertise in technology commercialisation, market development and scaling innovation-driven businesses.

Jenesys — Leadership
Jaspal Sarai
Jaspal Sarai
Chief Executive Officer

Technology entrepreneur and business leader with experience developing strategic partnerships, market-entry initiatives and commercial opportunities across global technology and defence sectors. Brings extensive experience in customer engagement and business development.

Nishq Ravindranath
Nishq Ravindranath
Chief Technology Officer

AI and autonomous systems specialist with expertise in advanced machine learning, distributed intelligence and mission-critical technology development. Experienced in designing scalable software architectures that support resilient, intelligent systems across defence, aerospace and real-world operational environments.

Awards & Recognition

National recognition for the Jenesys technology leadership this acquisition brings across.

Australian AI Awards 2024 winner seal
AI Leader of the Year — SME
Australian AI Awards 2024 Winner
Avalon 2025 Innovation Pitchfest and Awards winner seal
Emerging Technology Award
Avalon 2025 Innovation Pitchfest Winner
Australian Space Awards 2025 finalist seal
Innovator of the Year
Australian Space Awards 2025 Finalist
Australian AI Awards presentation
Avalon 2025 Innovation Pitchfest award presentation

Corporate Snapshot

4DS Memory Limited (ASX:4DS) is seeking to raise A$5 million to support the proposed acquisition of Jenesys, bringing together advanced memory, Edge AI and autonomous systems technologies.

As at 30 June 2026
ASX Code4DS
Share Price (last traded)11.1c
Shares on Issue2.06B
Unlisted & Listed Options275M
Market Capitalisation (AUD)2$22.66M
Cash in Bank (AUD)3$6.98M
Top 20 Shareholders24.79%
ASX:4DS — 12 Month Price
$0.070 $0.060 $0.050 $0.040 $0.030 $0.020 $0.010 APR JUL OCT 2026 APR
Capital Position
Current Position (Pre-Raise)As at 30 June 2026 1, 3
Pro Forma Position (Post A$5M Raise)(Indicative) 4
ASX Code
4DS
4DS
Share Price / Raise Price
1.1c1
1c (raise price)
Shares on Issue
2.06B
2.55B
New Shares to be Issued (on acquisition)
–
500M
Unlisted & Listed Options
275M
275M
Market Capitalisation (AUD)2
$22.66M
$30.5M
Cash (AUD)
$6.98M3
$11.98M5
Capital Raised (AUD)
–
$5.0M

1 Last traded price prior to voluntary suspension.

2 Market capitalisation based on last traded price prior to voluntary suspension.

3 Cash in Bank as at 30 June 2026 (unaudited).

4 Pro forma position is indicative and assumes completion of a A$5 million capital raising at 1c per share (before costs) and does not reflect any application of funds to acquisition consideration, transaction costs, working capital or other uses.

5 Indicative pro forma cash before costs and any application of funds.

Initialising
Third-party market data — valuations as at the date shown, drawn from a named listed-drone index plus two hand-added listings. Not Jenesys performance, not a forecast, and not a competitive claim about any company shown.
4DS Memory Jenesys logo JENESYS
Next-Gen Edge AI Capability

Proven memory. Trusted autonomy.
One integrated platform.

Combining 4DS' proprietary ReRAM technology platform with Jenesys' autonomous coordination software to support the emerging autonomous defence ecosystem.

4DS MEMORY LIMITED (ASX:4DS) · WWW.4DSMEMORY.COM · JENESYS.COM.AU
Jenesys logoJENESYS
One coordination layer. Every domain.

From a single platform to a coordinated fleet — AI that runs on the platform itself, and a coordination layer that turns many platforms into one force. Sovereign, platform-agnostic, and built in Australia.

JASPAL@JENESYS.COM.AU· JENESYS.COM.AU· BUILT IN AUSTRALIA
Air · Land · Sea · Space — One Coordination Layer