UPDATE August 4th 2015:The ASHM Sub-Committee for Guidance on HIV Management Committee recommends that antiretroviral therapy should be initiated in all people with HIV, irrespective of CD4 count, taking into account the following principles:
- Antiretroviral therapy (ART) is recommended for all HIV-infected individuals, irrespective of CD4 count, to reduce the risk of disease progression.
- The decision to start ART should take into account both personal health benefits and risks, and reduction in transmission risk
- Clinicians should regularly discuss the current state of knowledge regarding when to start ART with all individuals with HIV who are not yet on treatment
- All decisions to start ART should be made by the individual with HIV, in consultation with their health care providers and on the basis that they are fully informed and supported in their decision making.
This recommendation is consistent with guidance from the US DHHS Panel, which has recently been upgraded to an A1 Recommendation (strong recommendation based on data from randomised controlled trials) due to the publication of the results of the START and TEMPRANO Trials.
Treatment as prevention
October 2017 - The final results of the HPTN052 randomised trial of early versus delayed treatment of HIV positive people in HIV serodiscordant heterosexual couples showed that early ARV treatment of the HIV positive partner was associated with a 93% reduction in transmission risk to the negative partner. The only cases of HIV transmission from a person on treatment occurred very soon (days to several weeks) after treatment initiation or after treatment failure when viral load was detectable 
Since the interim results of HPTN052 published in 2011  showed a dramatic reduction in HIV transmission associated with early HIV treatment in heterosexual couples, randomised trials of the treatment as prevention transmission are no longer ethical. However, recent data from two observational studies of homosexual male serodiscordant couples have provided strong evidence that treatment leading to undetectable viral load reduces transmission risk to negligible levels. The European PARTNER study  reported no HIV transmissions in 308 couple years of follow-up when 22,000 acts of condomless anal intercourse were reported and the HIV positive partner had undetectable viral load. Data from the Australian-Thai-Brazilian Opposites Attract study, reported no transmissions in close to 600 couple-years of follow up with over 12,000 condomless anal intercourse acts protected by undetectable viral load . Data from these two studies strongly suggest that anti-retroviral therapy leading to undetectable viral load reduces HIV transmission in serodiscordant homosexual couples to close to zero, even in the presence of sexually transmissible infections. .
There has never been a well-documented case of HIV transmission from a person with undetectable viral load. Education about the protective effect of undetectable viral load are now being widely used in HIV prevention campaigns.
 Cohen MS et al; New England Journal of Medicine. 2016; 1; 375(9):830-9.
 Cohen MS, et al. N England Journal of Medicine. 2011; 11; 365(6):493-505
 Rodger AJ, et al. JAMA. 2016; 12;316 (2):171-81
 Bavinton BR, et al. Presented at International AIDS Society Conference, Paris 2017.