AI Designs Viruses That Infect Bacteria, but Success Rate Is Only 5%
AI has, for the first time, designed complete virus genomes from scratch that can infect bacteria; the experimental validation success rate is only about 5%, far from the all-capable image portrayed in the media.
The video won't play here. Listen to the audio instead:
The argument · tap a timestamp to hear it
Vaccine switching as a natural control
The pitfall of the 'healthy vaccinee bias' is that vaccinated individuals tend to be more health-conscious than the unvaccinated, so comparing the two groups can mistake lifestyle differences for vaccine effects. This study exploits the 2017 US switch of the shingles vaccine from Zostavax to Shingrix: those vaccinated before the policy change got the old vaccine, and those after got the new one; both groups were actively vaccinated, theoretically eliminating the bias. However, it is still not a randomized trial; unmeasured factors such as socioeconomic status, smoking, and diet may still be unbalanced between the two groups, so it can only be considered a quasi-experiment.
— Vincent RacanielloWeak confounding can overturn vaccine benefit
How strong must unmeasured confounding be to overturn an observational conclusion? The E-value measures this. This vaccine study has an E-value of only 1.42, barely above 1, meaning a confounding factor that is not very strong and is associated with both 'which vaccine received' and 'cardiovascular outcome' could explain away the observed risk reduction. An E-value of 3.4 or even 5 would indicate a robust conclusion; 1.42 is far from robust. The authors themselves admit that randomized clinical trials are needed for confirmation, and this observation should not be taken as evidence.
— Brianne BarkerAI creates functional virus genomes for the first time
The researchers used genomic language models EVO1 and EVO2 to write phage genomes: first letting the models read massive public DNA sequences, then fine-tuning on PhiX174 to generate thousands of candidate sequences. Back in the lab, only about 5% of the products could actually infect E. coli. This is the first time AI has designed complete genomes from scratch that are proven functional. If the media summarizes it as 'AI directly creates viruses,' they miss a key point: the models only see sequences, with no experimental annotations telling them which base does what. The 5% hit rate shows how ignorant they still are about how sequence determines function.
— Vincent RacanielloReplication origin is the model's most conserved memory
The prompt to the model only needed the first 4 to 9 nucleotides of the PhiX174 genome, and the accuracy of the generated results increased significantly. The authors found that this unremarkable sequence is part of the replication origin and is 100% conserved across all genomes generated by the model. It acts like a password saying 'I am a real virus': the model locks the replication origin as a core element, allowing evolution elsewhere. This also explains why such a short prompt is effective—the key is not prompt length but whether it hits a functional Achilles' heel.
— Kathy SpindlerAI swaps one gene and adds 58 mutations itself
EVO36's design is not a full rewrite but a gene swap: replacing PhiX174's J gene with the shorter J protein from phage G4. In the past, human manual transplantation of the same kind failed to produce infectious particles; but EVO36, given by AI, infects normally because it accumulated an additional 58 mutations across the genome. These mutations appear to be compensatory, allowing the foreign J protein to re-adapt to the local environment. Humans might not think to change 58 positions at once; this is where generative models surpass single-point editing.
— Vincent RacanielloPhage cocktails target drug-resistant bacteria
Natural PhiX174 hits a wall against already resistant bacteria, failing to invade even after five passages; a mixture of 16 AI-synthesized phages infects the resistant strain within one or two passages. The breakthrough against resistance does not rely on a single phage but on several synthetic phages recombining during replication, accumulating new mutations, and collectively breaching the barrier. This result suggests that phage therapy should use cocktails of multiple synthetic phages rather than betting on a single strain.
— Vincent RacanielloDon't frame AI-made viruses as an urgent threat
For the study generating PhiX174, a commentary used the word 'urgent' to describe its biosecurity implications. Vincent directly opposed this wording on the show: the authors have already shown through reproducible experiments that such research can be done safely, so the real task is not to sound the alarm but to start building a safety framework. He also stressed that reconstructing viruses is not science fiction as long as the sequence is available, which is precisely why we need frameworks rather than panic.
— Vincent RacanielloNot vaccinating is often just not knowing
Discussing the unvaccinated, the article raises an often-overlooked mechanism: some people do not get vaccinated not out of religious or ideological opposition but simply because they do not know they should. Treating all vaccine hesitancy as a 'belief issue' misdirects policy—for these people, the key is not debate or persuasion but information outreach so they first learn the vaccine exists.
In their own words · checked verbatim
It's only sequence. There is no biological knowledge associated with these sequences. But these are sequences from things that work. So the assumption is that you can learn from it.
Vincent Racaniello40:15
But it's not right. The genome sequence is everything. And if you have enough sequences, you could build a virus.
Vincent Racaniello1:09:46
I think it's also kind of cool in terms of helping us with basic discovery in that we might not have known that putting the J protein in in this way would have worked.
Vincent Racaniello1:10:46
It's not a religious thing. It's just they don't know that they should.
In his case, they said, well, your research is not of interest to the American public.
Figures
| Risk reduction for cardiovascular events in Shingrix group | 9% | 9:39 |
| E-value | 1.42 | 16:56 |
| EVO2 training data size | 9.3 trillion nucleotides | 40:15 |
| Efficiency of AI-generated sequences | about 5% (Evo 1: 5.3%, Evo 2: 6.9%) | 52:50 |
| Synthesized and functionally validated candidate genomes | 285 synthesized, 16 functional | 51:45 |
| Number of mutations EVO36 added for G4J protein | 58 | 56:55 |
| Drop in exam scores corresponding to using AI for homework | 25% | 1:23:27 |
Glossary
- E-value
- A measure of how strong unmeasured confounding would need to be to explain an observed association; the closer to 1, the more fragile the conclusion.
- healthy vaccinee bias
- Vaccinated individuals tend to be more health-conscious than the unvaccinated, causing observational studies to overestimate vaccine effectiveness.
- genomic language model
- A large model trained on DNA sequences as text, learning sequence patterns to generate new genomes.
- AS01 adjuvant
- The immune enhancer in the shingles vaccine Shingrix, thought to trigger nonspecific protective effects.
- compensation mutation
- An additional mutation that offsets the functional damage of another mutation, common in AI design.
- phage cocktail
- A therapy that mixes multiple phages to reduce the risk of bacterial resistance.
How to listen
Tech journalists and investors at the AI-biology intersection who are skeptical of 'AI-generated virus' reports, as well as evidence-based medicine practitioners who need to evaluate observational vaccine studies.
Readers not interested in vaccine confounding methods can jump to the AI phage section at 38 minutes.