What biotechnology covers
Biotechnology is older than its name: brewing, bread and cheese are all controlled fermentation. The modern field adds two things — the ability to read and write DNA, and the ability to manufacture biological molecules at industrial scale. Today it spans medicine (drugs, vaccines, cell and gene therapies), agriculture (engineered crops, biological pest control), industrial processes (enzymes, biofuels, precision-fermented proteins) and environmental applications (bioremediation, biosensors). The unifying idea: biology is a manufacturing platform with self-assembling, self-repairing products.
Medicine remains the largest and most closely watched segment, and its centre of gravity has shifted from small synthetic molecules to biologics — proteins produced by living cells — and to therapies that edit or replace a patient's own cells. That shift is why manufacturing and delivery, not just discovery, decide whether a therapy reaches patients.
The core technologies
Genetic engineering and CRISPR. The 2020 Nobel Prize in Chemistry recognised CRISPR-Cas9, the bacterial immune system repurposed into a programmable gene editor. A short guide RNA steers a cutting enzyme to any chosen DNA sequence. Successors (base editing, prime editing) cut less and rewrite more precisely. The first CRISPR-based medicine — a sickle-cell gene-edited therapy — was approved in late 2023, a decade after the technique's 2012 demonstration. Editing cells outside the body, verifying them, and re-infusing them is now a working clinical pattern.
mRNA platforms. mRNA vaccines deliver instructions rather than the antigen itself, so the body briefly manufactures the target protein and trains immunity against it. Designing a new mRNA vaccine means changing the instruction sequence, not the factory — the speed advantage the world saw in 2020–2021. The platform is being tested against flu, respiratory viruses and cancers, with the same core limitation each time: delivery. Getting RNA into the right cells (usually via lipid nanoparticles) remains the hard engineering problem.
Cell and gene therapy. CAR-T therapy reprograms a patient's own T cells to recognise cancer and is approved for several blood cancers; gene therapy replaces or corrects faulty genes for a growing list of inherited disorders. These are one-time, high-cost treatments that demand specialist manufacturing — the frontier where clinical success meets health-economics reality.
Sequencing and synthesis. Reading a human genome cost billions in the early 2000s and hundreds of dollars today; writing DNA is nearly as routine for short sequences. Cheap read/write capability is the quiet enabler behind everything above — including the AI applications below, because models are only as good as the data high-throughput sequencing produces.
How AI entered the lab
Proteins are chains of amino acids whose function is determined by the three-dimensional shape they fold into; predicting that shape from sequence was a fifty-year grand challenge. DeepMind's AlphaFold effectively solved the prediction problem for known sequences, and its predicted structures — alongside experimentally determined ones — are publicly available through community databases. The 2024 Nobel Prize in Chemistry honoured that work and, alongside it, computational protein design: the same logic run in reverse to invent proteins that do not exist in nature. For drug discovery, machine learning now screens candidate molecules, predicts their properties and helps design synthesis routes — compressing the search phase that traditionally consumed the most time. What it does not compress is the validation phase: candidate drugs still must be synthesised, tested for safety, and run through clinical trials measured in years.
For researchers and students, the practical entry points are documented in our AI tools for biology guide — literature search and source-grounded assistants listed with verified plans and limits — and the broader toolkit in AI research tools. For context on where the methods themselves are heading, our AI research coverage and the machine-learning desk follow the underlying model work.
What the field can and cannot do yet
The capabilities are real: approved gene-editing therapies, validated mRNA platforms, AI-structured proteins guiding drug programmes. The limits are equally real. Delivery — getting therapies into the right cells — remains the bottleneck for genetic medicine. Manufacturing biologics is slow, expensive and stringently regulated for good reason. Clinical trials take years and most candidates fail, which is the system working as designed. And access is uneven: approved therapies can cost millions of dollars per patient, raising the distribution questions that accompany every powerful technology. Treat single-study headlines accordingly — a promising result in mice or a small cohort is the start of a decade, not the end of one.
Governance and ethics
Biotechnology is governed by published, reviewable processes: clinical-trial registration, regulator approval (the FDA, the European Medicines Agency and national equivalents), and post-market monitoring. The sharpest ethical line is human germline editing — changes heritable by future generations — which the World Health Organization's reviews treat as off-limits under current conditions, after the 2018 embryo-editing incident showed why. For readers evaluating biotech claims, the strongest habit is the one we apply across this publication: find the primary source (approval letter, trial registration, peer-reviewed paper) before trusting the summary. Our critical-thinking guide walks through that process step by step.
Frequently asked questions
- What is biotechnology in simple terms?
- Biotechnology is the use of living organisms or their components — cells, enzymes, DNA — to make useful products. Insulin produced by engineered bacteria, enzymes used in detergents and cheese-making, crops engineered for pest resistance, and mRNA vaccines are all biotechnology. The common thread is borrowing biology's own machinery and directing it at a human problem.
- What is CRISPR and why does it matter?
- CRISPR-Cas9 is a gene-editing system adapted from a bacterial immune defence. A guide RNA steers the Cas9 enzyme to a chosen DNA sequence and cuts it, letting researchers disable or replace a gene. It matters because gene editing went from bespoke and expensive to routine and cheap in a few years — and because the first CRISPR-based therapies, such as the sickle-cell treatment approved in late 2023, turned it into real medicine.
- How is AI used in biotechnology?
- The headline example is protein structure prediction: DeepMind's AlphaFold predicted structures for hundreds of millions of proteins, work recognised with the 2024 Nobel Prize in Chemistry, and experimental structures are now public through databases like the Protein Data Bank. Beyond that, machine learning screens drug candidates, designs new proteins, reads medical images, and helps interpret genomic data. The honest caveat: AI narrows and accelerates the search, but drugs still need synthesis, safety testing and clinical trials.
- What is mRNA technology?
- mRNA (messenger RNA) vaccines deliver molecular instructions that let the body's own cells temporarily build an antigen — a harmless piece of a pathogen — which the immune system then learns to attack. The COVID-19 vaccines validated the platform at scale: faster to design than traditional vaccines because only the instruction molecule changes, not the manufacturing process. The same platform is now being tested for other viruses and for therapeutic uses.
- Are gene therapies widely available?
- Only for a growing but still narrow set of conditions. Several cell and gene therapies are approved — CAR-T cell therapies for certain blood cancers, gene therapies for specific inherited disorders — and each is expensive and complex to deliver. Most remain for serious conditions with few alternatives, and availability varies by country and health system. The FDA's cellular and gene therapy pages list what is actually approved.
- What are the main risks and ethical concerns?
- The field's risks range from the technical (off-target edits, immune reactions, manufacturing consistency) to the societal: equitable access to very expensive therapies, privacy of genomic data, and a bright ethical line around editing human embryos for reproduction, which the WHO and most national regulators prohibit or restrict. Responsible use is as much governance as science — which is why approval decisions are published and reviewable.